{"id":13819,"date":"2024-06-10T13:06:48","date_gmt":"2024-06-10T16:06:48","guid":{"rendered":"https:\/\/www.est.ufmg.br\/portal\/?page_id=13819"},"modified":"2026-04-10T18:05:20","modified_gmt":"2026-04-10T21:05:20","slug":"seminarios-do-dest","status":"publish","type":"page","link":"https:\/\/www.est.ufmg.br\/portal\/seminarios-do-dest\/","title":{"rendered":"Semin\u00e1rios & Videos do DEST"},"content":{"rendered":"\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Semin\u00e1rios do DEST<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

ANO DE 2026 – 1\u00ba SEMESTRE<\/strong><\/p>

17\/04\/2026 \u00e0s 13:30hs \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Luiz Henrique Duczmal (DEST-UFMG)
<\/b><\/p>

T\u00edtulo:<\/strong> Particle Manifold Metropolis-adjusted Langevin Algorithms<\/p>

Resumo:<\/b><\/p>

We propose a tree-spatial scan statistic that combines Kulldorff\u2019s circular scan method\u00a0for detecting spatial clusters and the tree-based scan statistic algorithm for data mining. We feed the tree-based scan algorithm with spatial information of events, which\u00a0are naturally arranged hierarchically. The tree-based scan statistic then examines all\u00a0possible branches of the tree to identify the branch where the associated probability of cases is higher than expected under the hypothesis of event homogeneity. The\u00a0algorithm was evaluated through simulations with hypothetical scenarios considering spatial and hierarchical structures, showing good performance in detecting these\u00a0structures. The tree-spatial scan method was applied to infant mortality data for the\u00a0Brazilian state of Rio de Janeiro in 2016. The proposed method identified a set of\u00a0municipalities in Rio de Janeiro where a branch of diseases had a significantly higher\u00a0number of deaths than expected under the homogeneity hypothesis.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t\t\t\t\t\n\t\t\t\t\t\n\t\t\t\t\t\tYoutube<\/span>\n\t\t\t\t\t\t<\/i>\t\t\t\t\t<\/a>\n\t\t\t\t<\/span>\n\t\t\t\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

V\u00eddeos mais Recentes do Canal<\/h3>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Semin\u00e1rios Online sobre a COVID-19\u200b<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
\u00a0<\/address>
Os semin\u00e1rios foram organizados pelos professores Luiz H. Duczmal e Glaura C. Franco.<\/address><\/address>

\"\"Os semin\u00e1rios est\u00e3o dispon\u00edveis em nosso canal do Youtube “Video conferencia do DEST”.<\/a><\/p><\/address>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t

\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t\"\"\t\t\t\t\t\t\t\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t
\n\t\t\t
<\/div>\n\t\t<\/div>\n\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t

Lista de Semin\u00e1rios do DEST<\/h2>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

ANO DE 2026 – 1\u00ba SEMESTRE<\/strong><\/p>\n

\n

10\/04\/2026 \u00e0s 13:30hs \u2013 Local: sala 2076 – ICEx<\/strong><\/p>\n

Uriel Moreira Silva (DEST-UFMG)
<\/b><\/p>\n

T\u00edtulo:<\/strong> Particle Manifold Metropolis-adjusted Langevin Algorithms<\/p>\n

Resumo: <\/b>In this work we propose a manifold version of the particle Metropolis-adjusted Langevin Algorithm (pMALA) of Nemeth et al. (2016)’s for parameter inference in State Space Models (SSM), and which we name particle Manifold Metropolis-adjusted Langevin Algorithm (pmMALA). Our method is a modification of pMALA that uses low-variance Hessian estimates of the log-target density as a metric tensor in the context of Riemannian Manifold Hybrid Monte Carlo (RMHMC) algorithms. A key ingredient in order to ensure proper convergence of RMHMC methods is that the metric tensor is positive definite, and here we satisfy this condition by employing adaptive step-size selection methods that originate from the nonlinear optimization literature and that were recently adapted for RMHMC. The end result is a method that does not require manual tuning of hyperparameters or pre-conditioning of the covariance matrix based on a pilot run, and that can achieve optimal scaling under relatively weak conditions. We illustrate pmMALA’s performance using both synthetic and real data, and show that it can obtain substantial inferential gains over conventional pMCMC and pMALA even in challenging nonlinear settings.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

27\/03\/2026 \u00e0s 13:30hs \u2013 Local: sala 2076 – ICEx<\/strong><\/p>\n

Gabriel Oliveira Assun\u00e7\u00e3o (DEST-UFMG)
<\/b><\/p>\n

T\u00edtulo:<\/strong> Is Augmentation Effective in Improving Prediction in Imbalanced Datasets?<\/p>\n

Resumo: <\/b>Neste semin\u00e1rio, abordaremos o problema de desbalanceamento de classes em modelos de machine learning, discutindo o uso de t\u00e9cnicas como o oversampling e suas alternativas. Em particular, ser\u00e1 explorado como o ajuste do limiar de decis\u00e3o do classificador pode influenciar o desempenho do modelo, muitas vezes de forma equivalente ao uso de dados sint\u00e9ticos. A proposta \u00e9 promover uma reflex\u00e3o sobre diferentes estrat\u00e9gias para lidar com esse tipo de problema na pr\u00e1tica.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

13\/03\/2026 \u00e0s 13:30hs \u2013 Local: sala 2027 (excepcionalmente, nesta semana) – DEST\/ICEx<\/strong><\/p>

Elias Teixeira Krainsk (King Abdullah University of Science and Technology, Ar\u00e1bia Saudita)<\/b><\/p>

T\u00edtulo:<\/strong> A graph-based framework for modeling correlation matrices that integrates expert-informed priors.<\/p>

Resumo:<\/b> The correlation coefficient is an important summary of the dependency between two variables. When studying several variables, the correlation matrix summarizes the marginal linear relationships among them. In data learning a correlation matrix some challenges arise, including rapid growth in the number of parameters. We address this challenge by proposing an approach to model correlation matrices using a graph to represent conditional dependency assumptions. We show that our approach requires as many parameters as the number of edges in the graph. We further leverage expert knowledge by introducing a prior that penalizes divergence from a base correlation matrix. When combined, it provides an informative model-based prior for correlation matrices. We will illustrate our approach with an application.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

12\/12\/2025 \u00e0s 13:30hs \u00e0s 14:30hs \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Syed Muhammad Arsalan (Doutorando em Estat\u00edstica-UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Online Updating of Survival Analysis<\/p>

Resumo:<\/b>\u00a0When large amounts of survival data arrive in streams, conventional estimation methods become computationally infeasible since they require access to all observations at each accumulation point. We develop online updating methods for carrying out survival analysis under the Cox proportional hazards model in an online-update framework. Our methods are also applicable with time-dependent covariates. Specifically, we propose online-updating estimators as well as their standard errors for both the regression coefficients and the baseline hazard function. Extensive simulation studies are conducted to investigate the empirical performance of the proposed estimators. A large colon cancer dataset from the Surveillance, Epidemiology, and End Results program and a large venture capital dataset with time-dependent covariates are analyzed to demonstrate the utility of the proposed methodologies. Supplemental files for this article are available online. This presentation is based on the paper by Jing Wu, Ming-Hui Chen, Elizabeth D. Schifano & Jun Yan (Journal of Computational and Graphical Statistics, DOI: https:\/\/doi.org\/10.1080\/10618600.2020.1870481)<\/p>

Orientador:<\/strong> F\u00e1bio Nogueira Demarqui (DEST-UFMG)
Coorientador:<\/strong> Marcos Oliveira Prates (DEST-UFMG)
Banca:<\/strong> Vin\u00edcius Diniz Mayrink Duarte (DEST-UFMG)<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t

\n\t\t\t\t\t\t
\n\t\t\t\t\t
\n\t\t\t
\n\t\t\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

SEMIN\u00c1RIO 1<\/b><\/p>

05\/12\/2025 \u00e0s 13:30hs \u00e0s 14:30hs \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Fl\u00e1vio dos Reis Moura (Doutorando em Estat\u00edstica-UFMG)<\/b><\/p>

T\u00edtulo:<\/strong> Modelos ARA e ARI de reparo imperfeito: estima\u00e7\u00e3o, avalia\u00e7\u00e3o de ajuste e predi\u00e7\u00e3o de confiabilidade<\/p>

Resumo:<\/b>\u00a0Esta apresenta\u00e7\u00e3o discute, de forma did\u00e1tica e aprofundada, os modelos de reparo imperfeito conhecidos como ARA (Arithmetic Reduction of Age) e ARI (Arithmetic Reduction of Intensity), aplicados \u00e0 an\u00e1lise de sistemas repar\u00e1veis. A partir do artigo \u201cARA and ARI Imperfect Repair Models: Estimation, Goodness-of-Fit and Reliability Prediction\u201d, ser\u00e3o apresentados os conceitos de idade virtual e redu\u00e7\u00e3o de intensidade ap\u00f3s o reparo, permitindo descrever situa\u00e7\u00f5es intermedi\u00e1rias entre reparo perfeito (AGAN) e reparo m\u00ednimo (ABAO). Inicialmente, o Processo de Lei de Pot\u00eancia (PLP) \u00e9 introduzido como modelo de base para a intensidade de falhas, com destaque para sua fun\u00e7\u00e3o intensidade, risco acumulado e fun\u00e7\u00e3o cumulativa m\u00e9dia (MCF). Em seguida, s\u00e3o apresentados os modelos ARA e ARI como extens\u00f5es do PLP, detalhando suas formula\u00e7\u00f5es matem\u00e1ticas, interpreta\u00e7\u00e3o f\u00edsica e diferen\u00e7as no tratamento do efeito do reparo. A apresenta\u00e7\u00e3o abordar\u00e1 os procedimentos de estima\u00e7\u00e3o por m\u00e1xima verossimilhan\u00e7a, o c\u00e1lculo de medidas de confiabilidade p\u00f3s-reparo e os m\u00e9todos de avalia\u00e7\u00e3o de ajuste, incluindo crit\u00e9rios de informa\u00e7\u00e3o (AIC, BIC) e diagn\u00f3sticos baseados em res\u00edduos e QQ-plots. Tamb\u00e9m ser\u00e3o discutidos resultados de aplica\u00e7\u00e3o a dados reais de falhas de caminh\u00f5es de uma mineradora brasileira, bem como a utiliza\u00e7\u00e3o dos modelos para predi\u00e7\u00e3o de confiabilidade e apoio \u00e0 defini\u00e7\u00e3o de pol\u00edticas de manuten\u00e7\u00e3o preventiva.<\/p>

Baseado no artigo\u00a0 “ARA and ARI Imperfect Repair Models: Estimation, Goodness-of-Fit and Reliability Prediction “, dos autores\u00a0 \u00a0Maria Lu\u00edza Guerra de Toledo, Marta A. Freitas, Enrico A. Colosimo, Gustavo L. Gilardoni<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

SEMIN\u00c1RIO 2<\/b><\/p>

05\/12\/2025 \u00e0s 14:40hs \u00e0s 15:40hs \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Eriky S. Gomes (Doutorando em Estat\u00edstica-UFMG)<\/b><\/p>

T\u00edtulo:<\/strong> Change point estimation of service rate in M\/M\/1\/m queues: A Bayesian approach<\/p>

Resumo:<\/b>\u00a0As filas M\/M\/1\/m modelam sistemas de filas com servidor \u00fanico e capacidade limitada. Nesse trabalho, admite-se a exist\u00eancia de um ponto de mudan\u00e7a na distribui\u00e7\u00e3o do tempo de servi\u00e7o, devido, por exemplo, \u00e0 altera\u00e7\u00e3o na efici\u00eancia do servi\u00e7o ou nas caracter\u00edsticas dos usu\u00e1rios do sistema. A detec\u00e7\u00e3o dos pontos de mudan\u00e7a \u00e9 feita por meio da constru\u00e7\u00e3o da verossimilhan\u00e7a e das distribui\u00e7\u00f5es a posteriori obtidas da observa\u00e7\u00e3o do n\u00famero de usu\u00e1rios presentes no sistema imediatamente ap\u00f3s a conclus\u00e3o de cada servi\u00e7o. Simula\u00e7\u00f5es de Monte Carlo s\u00e3o realizadas para a an\u00e1lise de desempenho dos estimadores propostos. Tamb\u00e9m ser\u00e3o discutidas propostas de estima\u00e7\u00e3o via N\u00facleo Estimador e Product Partition Models nesse contexto. This presentation is based on the paper by S.K.Singh, G.M.B.Cruz, F.R.B.Cruz, Change point estimation of service rate in M\/M\/1\/m queues: A Bayesian approach, Applied Mathematics and Computation, vol.465, 128423, 2024. Doi: https:\/\/doi.org\/10.1016\/j.amc.2023.128423}{10.1016\/j.amc.2023.128423.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

14\/11\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Ana J\u00falia Alves C\u00e2mara (Departamento de Estat\u00edstica – UFES)<\/strong><\/p>

T\u00edtulo:<\/strong> Modelagem de processos de contagem e deriva\u00e7\u00f5es: Aplica\u00e7\u00f5es em dados ambientais e epidemiol\u00f3gicos.<\/p>

Resumo: <\/b>Neste semin\u00e1rio, ser\u00e3o apresentados os resultados centrais de uma linha de pesquisa dedicada \u00e0 modelagem estat\u00edstica de s\u00e9ries temporais de contagens na epidemiologia ambiental. A investiga\u00e7\u00e3o parte da necessidade de aprimorar a an\u00e1lise de processos n\u00e3o Gaussianos e autocorrelacionados que descrevem a din\u00e2mica de morbidade e mortalidade associada \u00e0 exposi\u00e7\u00e3o a poluentes atmosf\u00e9ricos. O primeiro estudo prop\u00f5e o modelo GAM-ARMA, que combina fun\u00e7\u00f5es splines a uma estrutura autorregressiva de m\u00e9dias m\u00f3veis, permitindo capturar simultaneamente n\u00e3o linearidades e depend\u00eancia temporal. O segundo integra o modelo GLARMA a diferentes abordagens de bootstrap, aprimorando a infer\u00eancia do risco relativo em s\u00e9ries com poucas observa\u00e7\u00f5es. O terceiro introduz um modelo GLARMA robusto, baseado em M-estimadores, com o objetivo de reduzir o impacto de observa\u00e7\u00f5es at\u00edpicas em covari\u00e1veis ambientais. As aplica\u00e7\u00f5es com dados das regi\u00f5es metropolitanas de Belo Horizonte\/MG e Vit\u00f3ria\/ES demonstram ganhos expressivos em ajuste, estabilidade e validade inferencial. Em conjunto, os trabalhos ampliam o uso dos modelos GLARMA, articulando robustez, flexibilidade e rigor estat\u00edstico na an\u00e1lise de fen\u00f4menos complexos de sa\u00fade e meio ambiente.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

07\/11\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof. Dani Gamerman (Departamento de M\u00e9todos Estat\u00edsticos – IM-UFRJ)<\/b><\/p>

T\u00edtulo:<\/strong> Dividir para conquistar: Infer\u00eancia Bayesiana exata em estat\u00edstica espacial.<\/p>

Resumo:\u00a0<\/b>A estat\u00edstica bayesiana cresceu consideravelmente nas \u00faltimas d\u00e9cadas, levando a modelos com complexidade crescente. Esse crescimento foi acompanhado pela necessidade de aproxima\u00e7\u00f5es, \u00e0s quais nos acostumamos. Algumas delas s\u00e3o apenas para extrair informa\u00e7\u00e3o da distribui\u00e7\u00e3o a posteriori, mas outras aproxima\u00e7\u00f5es s\u00e3o causadas pela incapacidade de lidar com os modelos conforme propostos. Um exemplo importante \u00e9 fornecido por padr\u00f5es de pontos (PP). Essa estrutura de dados \u00e9 comumente encontrada em an\u00e1lises espaciais, onde a estimativa da intensidade da ocorr\u00eancia dos pontos \u00e9 o principal interesse em muitos cen\u00e1rios usuais. No entanto, a fun\u00e7\u00e3o de verossimilhan\u00e7a para intensidades n\u00e3o param\u00e9tricas n\u00e3o est\u00e1 dispon\u00edvel analiticamente. Aproxima\u00e7\u00f5es s\u00e3o geralmente aplicadas, induzindo vieses e perdas em todos os procedimentos inferenciais baseados em verossimilhan\u00e7a. Essa inefici\u00eancia \u00e9 herdada por todos os modelos que cont\u00eam componentes PP. Processos gaussianos (GP) usados para induzir suavidade na intensidade trazem ainda mais complexidade. Esta palestra abordar\u00e1 essas complica\u00e7\u00f5es e propor\u00e1 procedimentos exatos para remediar a situa\u00e7\u00e3o. Esses procedimentos s\u00e3o baseados em aumento de dados com processos latentes e evitam aproxima\u00e7\u00f5es de modelos. Essa ideia gera uma ampla plataforma para lidar com modelos tendo componentes de PP sem comprometer sua integridade. Exemplos incluem:<\/p>

1) o uso de regress\u00f5es flex\u00edveis em PP,<\/p>

2) geoestat\u00edstica com amostragem preferencial,<\/p>

3) an\u00e1lise de dados de apenas-presen\u00e7a em Ecologia, e<\/p>

4) tratamento de fun\u00e7\u00f5es de intensidade que n\u00e3o s\u00e3o suaves.<\/p>

Resultados de testes com dados sint\u00e9ticos, compara\u00e7\u00f5es com alternativas e aplica\u00e7\u00f5es em dados reais de diversas \u00e1reas da Ci\u00eancia s\u00e3o apresentados para ilustrar as diferentes solu\u00e7\u00f5es adotada nos exemplos acima. Quest\u00f5es computacionais associadas ao custo de GP, \u00e0 paraleliza\u00e7\u00e3o e \u00e0 gera\u00e7\u00e3o de software s\u00e3o brevemente abordadas.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

31\/10\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Profa. Ilka Afonso Reis<\/b>\u00a0(DEST\/UFMG)<\/b><\/p>

T\u00edtulo:<\/strong> Do planejamento \u00e0 an\u00e1lise: das alegrias e agruras de uma profissional em Estat\u00edstica na intera\u00e7\u00e3o com o mundo real.<\/p>

Resumo:\u00a0<\/b>Nesse semin\u00e1rio, proponho uma reflex\u00e3o cr\u00edtica sobre os desafios enfrentados na aplica\u00e7\u00e3o pr\u00e1tica da Estat\u00edstica em contextos n\u00e3o controlados,\u00a0<\/span>a partir da minha experi\u00eancia profissional. Por meio de casos reais, ser\u00e3o discutidas as tens\u00f5es entre o rigor t\u00e9cnico e as demandas pragm\u00e1ticas\u00a0<\/span>de diferentes \u00e1reas de atua\u00e7\u00e3o, bem como os dilemas \u00e9ticos envolvidos na modelagem dos dados, interpreta\u00e7\u00e3o e comunica\u00e7\u00e3o dos resultados.\u00a0<\/span>Pretendo tamb\u00e9m destacar os aspectos positivos da pr\u00e1tica estat\u00edstica, como a contribui\u00e7\u00e3o efetiva para decis\u00f5es informadas, o impacto social\u00a0<\/span>das an\u00e1lises bem conduzidas e a constru\u00e7\u00e3o de pontes entre saberes disciplinares. Destinada a estudantes, docentes e pesquisadores, a palestra\u00a0<\/span>convida \u00e0 valoriza\u00e7\u00e3o da Estat\u00edstica como ci\u00eancia aplicada, ressaltando a import\u00e2ncia da forma\u00e7\u00e3o cr\u00edtica, da flexibilidade metodol\u00f3gica e da\u00a0<\/span>sensibilidade contextual na atua\u00e7\u00e3o profissional.<\/span><\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t
Workshop <\/strong>
Recent Advances in Complex Data Modeling<\/strong><\/h5>

DEST- UFMG- Room 2076<\/strong>
Friday, October 24, 2024 at 13h30<\/strong><\/p>

Reducing Dimensionality in Covariate-Informed Random Partition Models<\/strong><\/p>

Raffaele Argiento (Universit\u00e0 degli Studi de Bergamo, It)<\/strong><\/p>

Covariate-informed models for clustering, such as the Product Partition Model with covariates (PPMx), have proven effective in incorporating auxiliary information to improve clustering performance. However, in high-dimensional settings, selecting a low-dimensional subset of covariates becomes a central challenge: including irrelevant variables can distort clustering by overpowering the response signal and leading to overly granular or uninterpretable partitions. In this work, we introduce a novel approach based on Covariate Subset Selection (CSS). Although originally developed within computer science, CSS has recently been shown to admit a statistical formulation in which no specific assumptions are required for the selected covariates, while a multivariate regression model is specified for the remaining covariates, conditional on the selected subset.
Building on this foundation, we assign a Bayesian mixture model to the joint distribution of the response and the selected covariates. We show that the resulting model corresponds to a PPMx that automatically performs CSS as part of the clustering mechanism. This unified framework offers several advantages: it is probabilistically coherent, computationally efficient, and retains desirable properties such as Kolmogorov consistency, all while avoiding the complexities associated with reversible jump MCMC schemes. Simulation studies and real-data applications demonstrate the robustness and practical effectiveness of the proposed approach.<\/p>

Advances in Spatial Statistics for Large-Scale and Complex Domains<\/strong><\/p>

Marcos Oliveira Prates (DEST-UFMG)<\/strong><\/p>

The proliferation of large-scale geospatial data from sources such as satellite remote sensing and cellular phone networks has created a need for new statistical methods capable of handling massive datasets and complex spatial domains, as classical techniques often face prohibitive computational burdens and restrictive assumptions. In this talk, I discuss recent advances that directly address some of these challenges, primarily through the development of a scalable model that reduces computational complexity from cubic to near-linear in the number of observations. Further, we explore some of its applications. Beyond scalability, progress has been made in tailoring methods for complex domains by defining a process using appropriate distance metrics. The synthesis of these scalable and geometrically aware methods empowers practitioners to extract meaningful insights from vast and intricate spatial data. Again, we revisit applications in other spatial domains. FAPEMIG and CNPq partially funded these works. This is a joint work with Carlos Gonz\u00e1les, Dipak K. Dey, Harvard Rue, Heitor Ramos, Lucas Godoy, Lucas Michelin, Jun Yan, and Zaida Quiroz.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

10\/10\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof. Fl\u00e1vio Bambirra Gon\u00e7alves (DEST\/UFMG<\/span>)<\/b><\/p>

T\u00edtulo:<\/strong> Scalable Bernoulli Factory MCMC for Intractable Marginalised Posteriors.<\/p>

Resumo: <\/b>Bernoulli factory MCMC algorithms implement accept-reject Markov chains without explicit computation of acceptance probabilities, and are used to target posterior distributions associated with intractable likelihood models. Intractable likelihoods naturally arise in continuous -time models and mixture distributions, or from the marginalisation of a tractable augmented model. Bernoulli factory MCMC algorithms often mix better than alternatives that target a tractable augmented posterior. However, for a likelihood that factorizes over observations, we show that their computational performance typically deteriorates exponentially with data size. To address this, we propose a simple divide-and-conquer Bernoulli factory MCMC algorithm and prove that it has polynomial complexity of degree between 1 and 2, with the exact degree depending on the existence of efficient unbiased estimators of the intractable likelihood ratio. We demonstrate the effectiveness of our approach with applications to Bayesian inference in two intractable likelihood models, and observe respective polynomial cost of degree 1.2 and 1 in the data size.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

03\/10\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Profa. Camila Cristina Lara Prado\u00a0\u00a0(DEST\/UFMG<\/span>)<\/b><\/p>

T\u00edtulo:<\/strong> Timing Strategy of Commodity Managers.<\/p>

Resumo: <\/strong>The purpose of this research is to study whether commodity managers investors have the ability to time factor exposures. I utilize the methodology developed by Treynor and Mazuy (1966), and Henriksson and Merton (1981), and apply the four-factor commodity model of Blocher et al (2018). Specifically, I measure market timing, momentum timing, the high term (realized term premia for the commodities with above\u2010median basis), and low term (realized term premia for the commodities with below\u2010median basis) skills. These factors are chosen because each one, separately, captures a risk premium embedded in commodity futures. My results indicate that commodity managers\u2019 returns have some statistically significant market timing abilities. This means that many managers increase exposure to the nearest contract when the spot premium return is high and decrease exposure when the spot premium return is low. Momentum timing, high term timing, and low term timing are not observed. When looking at different strategies, technical managers demonstrate stronger market timing ability than fundamental managers.<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

26\/09\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Profa. Jussiane Nader Gon\u00e7alves<\/span> \u00a0(DEST\/UFMG<\/span>)<\/b><\/p>

T\u00edtulo:<\/strong> Bivariate mixed Poisson models: a flexible class of regressions for paired count data<\/span>.<\/p>

Resumo: <\/strong>In many fields, count variables often arise in a dependent manner, thus requiring a joint estimation method. Particularly in health care systems, providing reliable estimates of the number of events is crucial to evaluate health care costs. For instance, in the Brazilian health insurance system, actuaries seek to predict the number of doctor appointments and medical exams, as well as other groups of procedures, in order to establish fair premiums. Since a medical exam is performed when prescribed, it is naturally correlated with doctor appointments, making joint analysis of these variables essential.\u00a0Beyond this intrinsic correlation, unobserved heterogeneity also plays a role. In the previous example, such heterogeneity may arise from portfolio aging, socioeconomic and demographic factors, technological availability, or epidemiological conditions, among others. While some of these factors can be included in regression models as covariates, many relevant variables are difficult to measure, such as the quality of health service providers.\u00a0Motivated by these challenges, we propose a new class of regression models and study its properties and adequacy. Specifically, we introduce a flexible class of bivariate mixed Poisson (BMP) regression models, which incorporate an exponential-family (EF) distributed component to account for unobserved heterogeneity. This framework addresses overdispersion, a common feature of count data, and provides flexibility in terms of the correlation structure.<\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

12\/09\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof. Douglas Mateus da Silva (DEST\/UFMG<\/span>)<\/b><\/p>

T\u00edtulo:<\/strong> Exact Bayesian Geostatistics Under Preferential Sampling.<\/p>

Resumo:\u00a0<\/strong>Preferential sampling is a common feature in geostatistics and occurs when the locations are sampled based on information about the phenomena under study. In this case, point pattern models are commonly used as the probability law for the distribution of the locations. However, analytic intractability of the point process likelihood prevents its direct calculation. Many Bayesian (and non-Bayesian) approaches in non-parametric model specifications handle this difficulty with approximations, both to the model and to the computations required for drawing inference. Procedures to approximate the model lead to errors that are sometimes difficult to quantify and can lead to biased inference. This paper presents an approach for performing exact Bayesian inference for this setting without the need for model approximation. A qualitatively minor change on the traditional model is proposed to circumvent the likelihood intractability. This change enables the use of an augmented model strategy. Recent work on Bayesian inference for point pattern models can be adapted to the geostatistics setting and renders computational tractability for exact inference for the proposed methodology. Estimation of model parameters and prediction of the response at unsampled locations can then be obtained from the joint posterior distribution. Simulated studies showed good quality of the proposed model for estimation and prediction in a variety of scenarios. The performance of our approach is illustrated in the analysis of simulated and real datasets and also compares favourably against approximation-based approaches.\u00a0<\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

\u00a005\/09\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof. Vinicius Diniz Mayrink\u00a0(DEST\/UFMG<\/span>)<\/b><\/p>

T\u00edtulo:<\/strong>\u00a0Bessel regression and bbreg package for bounded data.<\/p>

Resumo:\u00a0<\/strong>Beta regression is a standard approach for modeling continuous bounded data, with limited competitors offering similar flexibility. The normalized inverse-Gaussian (N-IG) process, a Bayesian alternative to the Dirichlet process, has gained attention, yet its univariate\u00a0 distribution remains underexplored in classical inference. We introduce bessel regression, based on the univariate N-IG, as an alternative to beta regression. Parameters are estimated via an EM algorithm, and inference procedures are derived. A model selection criterion is proposed to distinguish between bessel and beta regressions. The R package bbreg implements both models and provides tools for model adequacy and selection. A simulation study assesses robustness under misspecification, and an empirical application illustrates\u00a0 the comparison. This work is a collaboration with Wagner Barreto-Souza (UCD, Ireland) and Alexandre B. Simas (KAUST, Saudi Arabia). The presenter, Vinicius D. Mayrink, acknowledges the support from CNPq, CAPES, and FAPEMIG.\u00a0<\/span><\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

29\/08\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof. Roger William Camara (DEST\/UFMG<\/span>)<\/b><\/p>

T\u00edtulo:<\/strong> Multirange percolation of words on the hypercubic lattice<\/p>

Resumo:\u00a0\u00a0<\/strong>We investigate the problem of percolation of words in a random environment. We independently assign each vertex a letter $0$ or $1$ according to Bernoulli random variables with mean $p$. The environment is the resulting graph obtained from an independent long-range bond percolation configuration on $\\Z^{d-1} \\times \\Z$, $d\\geq 3$, where each edge parallel to $\\Z^{d-1}$ has length one and is open with probability $\\epsilon$, while edges of length $n$ parallel to $\\Z$ areopen with probability $p_n$. We prove that if the sum of $p_n$ diverges, then for any $\\epsilon$ and $p$, there is a $K$ such that all words are seen from the origin with probability close to $1$, even if all connections with length larger than $K$ are suppressed.\u00a0<\/span><\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

11\/07\/2025 \u00e0s 10:00h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Profa. a Michelle Miranda (<\/span>University of Victoria\u00a0)<\/b><\/p>

T\u00edtulo:<\/strong> Mapping Long Memory in Resting-State fMRI: Age-Related Changes in the Hippocampus from the ADHD-200 Dataset<\/p>

Resumo: <\/strong>Resting-state functional MRI (fMRI) measures spontaneous brain activity while the subject is not engaged in any specific task, and has become a standard tool for studying intrinsic brain function. The res ulting time series often display long-range temporal dependence, which is not adequately captured by models widely used in neuroscience that assume independence or short-memory autoregressive structure.\u00a0In this talk, I introduce a statistical pipeline that models these dependencies using a long-memory (LM) framework, where autocorrelations decay according to a power law. LM parameters are estimated voxelwise via a wavelet-based Bayesian method, yielding spatial maps of temporal dependence. These maps are then projected onto a lower-dimensional space using a composite basis and related to subject-level covariates through regression. While resting-state studies typically focus on functional connectivity, it is hypothesized that the LM parameter captures complementary aspects of intrinsic brain activity. Applying this approach to the ADHD-200 dataset, we found that age in children is positively associated with the LM parameter in the hippocampus, after adjusting for ADHD sympt om severity and medication status. This work highlights the potential of long-memory modeling for studying developmental patterns in resting-state brain activity and provides a flexible framework for analyzing large-scale neuroimaging data.<\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

04\/07\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof. Louren\u00e7o Ribeiro (DEST\/UFMG)<\/b><\/p>

T\u00edtulo:<\/strong> Distance-based loss function for deep feature space learning of convolutional neural networks<\/p>

Resumo: <\/strong>Convolutional Neural Networks (CNNs) have been on the forefront of neural network research in recent years. Their breakthrough performance in fields such as image classification has gathered efforts in the development of new CNN-based architectures, but recently more attention has been directed to the study of new loss functions. Softmax loss remains the most popular loss function due mainly to its efficiency in class separation, but the function is unsatisfactory in terms of intra-class compactness. While some studies have addressed this problem, most solutions attempt to refine softmax loss or combine it with other approaches. We present a novel loss function based on distance matrices (LDMAT), softmax independent, that maximizes interclass distance and minimizes intraclass distance. The loss function operates directly on deep features, allowing their use on arbitrary classifiers. LDMAT minimizes the distance between two distance matrices, one constructed with the model\u2019s deep features and the other calculated from the labels. The use of a distance matrix in the loss function allows a two-dimensional representation of features and imposes a fixed distance between classes, while improving intra-class compactness.
<\/strong><\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

27\/06\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Profa. Denise Duarte (DEST\/UFMG)<\/b><\/p>

T\u00edtulo:<\/strong> Probabilistic Context Neighborhood model for lattices<\/p>

Resumo: <\/strong>We present the Probabilistic Context Neighborhood model designed for two-dimensional lattices as a variation of a Markov random field, assuming discrete values.\u00a0In this model, the neighborhood structure has a fixed geometry but a variable order of spatial dependence, which varies based on the values of its neighbors.\u00a0Our model extends the Probabilistic Context Tree model, originally applicable to one-dimensional space. It retains advantageous properties, such as representing the dependence\u00a0neighborhood structure as a graph in a tree format, facilitating an understanding of model complexity. Furthermore, we adapt the algorithm used to estimate the Probabilistic\u00a0Context Tree to estimate the parameters of the proposed model. We illustrate the accuracy of our estimation methodology through simulation studies. We apply the Probabilistic\u00a0Context Neighborhood model to spatial data from fires in the Pantanal, highlighting its practical utility. Joint with D\u00e9bora F. Magalh\u00e3es, Aline M. Piroutek, Caio Alves.<\/p>

\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\t\t\t\t

13\/06\/2025 \u00e0s 13:30h \u2013 Local: sala 2076 – ICEx<\/strong><\/p>

Prof.\u00a0 Frederico R. B. da Cruz<\/b><\/p>

T\u00edtulo:<\/strong> Algoritmos para Determina\u00e7\u00e3o de Tamanhos de Amostra em Estima\u00e7\u00e3o Bayesiana em Filas Markovianas de Servidor \u00danico<\/p>

Resumo:\u00a0<\/strong>Este trabalho, desenvolvido em colabora\u00e7\u00e3o com Eriky S. Gomes, doutorando do Programa de P\u00f3s-Gradua\u00e7\u00e3o em Estat\u00edstica da UFMG, e Saroja Kumar Singh, do Department of Statistics da Ravenshaw University (\u00cdndia), aborda o desenvolvimento de algoritmos bayesianos para determina\u00e7\u00e3o do tamanho de amostras no cl\u00e1ssico modelo de filas markovianas de servidor \u00fanico, ou filas M\/M\/1, na nota\u00e7\u00e3o de Kendall. Embora sejam modelos simples, essas filas possuem importantes aplica\u00e7\u00f5es pr\u00e1ticas, e um dos passos fundamentais para sua utiliza\u00e7\u00e3o \u00e9 a determina\u00e7\u00e3o do tamanho amostral necess\u00e1rio para a estima\u00e7\u00e3o intervalar de seus par\u00e2metros, especialmente da intensidade de tr\u00e1fego, definida como a raz\u00e3o entre as taxas de chegada e de atendimento. Propomos algoritmos que garantem, de forma eficiente, uma amplitude m\u00e9dia ou cobertura m\u00e9dia pr\u00e9-especificadas para o intervalo bayesiano da intensidade de tr\u00e1fego, utilizando como dados o n\u00famero de chegadas durante os tempos de atendimento, uma forma pr\u00e1tica de coleta. Simula\u00e7\u00f5es de Monte Carlo demonstram a efici\u00eancia e a efic\u00e1cia dos m\u00e9todos desenvolvidos.<\/p>

Maiores detalhes podem ser vistos em:<\/p>

GOMES, E. S.; CRUZ, F. R. B.; SINGH, S. K. Algorithms for Determination of Sample Sizes for Bayesian Estimations in Single-Server Markovian Queues. AMERICAN JOURNAL OF MATHEMATICAL AND MANAGEMENT SCIENCES, v. 42, p. 307-322, 2023. (http:\/\/dx.doi.org\/10.1080\/01966324.2023.2255316<\/a>)<\/p>

23\/05\/2025 \u00e0s 13:30h \u2013 Local: Via ZOOM<\/strong><\/p>

Link: https:\/\/us06web.zoom.us\/j\/85138086178?pwd=uNrGlY3ZHTfL43CvS6ukRBGYcBA1KZ.1<\/a>
ID da reuni\u00e3o: 851 3808 6178
Senha: 495662<\/strong><\/p>

Prof.\u00a0\u00a0<\/b>J Andr\u00e9s Christen(CIMAT-Mexico)<\/b><\/p>

T\u00edtulo:<\/strong> Dynamic survival analysis: modelling the hazard function via ordinary differential equations<\/p>

Resumo:<\/strong> We propose a general approach for parametrically modelling the hazard function using ODEs.Our proposal capitalizes on the extensive literature of ODEs which, in particular, allow for establishing basic rules or laws on the dynamics of autonomous ODEs which we use to model the hazard function. We present examples both when there is an analytic solution to the system of ODEs or where an ODE solver is required to obtain a numerical solution. From than, we present examples of Bayesian inference using known data bases. Two case studies using real data are presented to illustrate the use of the proposed approach and to highlight the interpretability of the corresponding models.
Joint work with Javier Rubio (UCL, UK)<\/p>

16\/05\/2025 \u00e0s 13:30h \u2013 Local: Via ZOOM<\/strong><\/p>

Link: https:\/\/us06web.zoom.us\/j\/89585065330?pwd=3SB50k20KtqwQGynDDAGhJoGrjcr7k.1<\/a>
ID da reuni\u00e3o: 895 8506 5330
Senha: 038165<\/strong><\/p>

Hedibert F. Lopes<\/strong>
Joint with: Pedro Lima, Carlos M. Carvalho, Andrew Herren<\/strong><\/p>

T\u00edtulo:<\/strong> Minnesota BART<\/p>

Resumo:<\/strong> Vector autoregression (VAR) models are crucial for forecasting and analyzing macroeconomic variables, serving as a fundamental tool for applied macroeconomists. Recent literature has explored nonparametric approaches, such as Bayesian additive regression trees (BART), which allow for flexibility without strong parametric assumptions; however, existing frameworks like the one proposed by Huber and Rossini (2022) do not adequately accommodate high-dimensional data or time dependency in the prior construction. This study enhances the literature by extending previous work to enable high-dimensional data analysis and variable selection through a sparsity-inducing Dirichlet hyperprior, as in Linero (2018) on the regression tree\u2019s splitting probabilities, while also proposing a prior that decrease the probability of splitting on variables that are have higher lags than smaller lags, similar to the approach taken by the Minnesota Prior. Empirical results show improvement compared to the baseline BART prior structure and a BVAR.<\/p>

25\/04\/2025 \u00e0s 13:30h \u2013 Local: Sala 2076 \u2013 ICEx\/UFMG<\/strong><\/p>

Marcelo Azevedo Costa\u00a0<\/b>(Department of Industrial Engineering\/UFMG)<\/p>

T\u00edtulo:<\/strong> Geographical Bayesian second stage analysis for operating efficiency of Brazilian electricity distribution system operators<\/p>

Resumo:<\/strong> Since 2011, the Brazilian electricity regulator has applied data envelopment analysis to estimate regulatory operating costs for distribution and transmission companies. Despite the availability of environmental or contextual variables, second-stage analysis has been avoided, primarily due to inconsistent statistical results, including estimated coefficients contrary\u00a0to technical evidence and significant changes in operating efficiencies for selected companies. Previous studies have shown that environmental adjustments are critical for companies\u2019 revenues operating in harsh environments in Brazil. Additionally, climate changes are affecting expenses with varying effects nationwide. To tackle this challenge, a second-stage model\u00a0in which changes in efficiencies are also affected by geographical location of companies is proposed. Coefficient constraints and multiple environmental variables are applied to estimate regulatory efficiencies of Brazilian Distributor System Operators forupcoming years. Results indicate maximum efficiency changes of +5.35% and an increase of 1.1% in the total regulatory OPEX if the proposed second stage is applied.<\/p>

11\/04\/2025 \u00e0s 13:30h \u2013 Local: Sala 2076 \u2013 ICEx\/UFMG<\/strong><\/p>

F\u00e1bio Nogueira Demarqui (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> A Modern Approach to Simulating Survival Data With the R Package rsurv.<\/p>

Resumo:<\/strong> In this talk we introduce the R package rsurv, aimed for general survival data simulation purposes. The package, which is available on CRAN at https:\/\/CRAN. R-project.org\/package=rsurv, was designed to make the simulation of survival data more intuitive and aligned with data modeling practices through the use of dplyr verbs. The proposed package allows the simulation of survival data from a wide range of regression models, including accelerated failure time (AFT), proportional hazards (PH), proportional odds (PO), accelerated hazard (AH), Yang and Prentice (YP), and extended hazard (EH) models. The package rsurv also stands out by its ability to generate survival data from an unlimited number of baseline distributions provided that an implementation of the quantile function of the chosen baseline distribution is available in R. Another nice feature of the package rsurv lies in the fact that linear predictors are specified via a formula-based approach, facilitating the inclusion of categorical variables and interaction terms. The functions implemented in the package rsurv can also be employed to simulate survival data with more complex structures, such as survival data with different types of censoring mechanisms, survival data with cure fraction, survival data with random effects (frailties), multivariate survival data, and competing risks survival data Keywords: Censoring, random data generation, survival regression models.<\/p>

21\/03\/2025 \u00e0s 13:30h \u2013 Local: Sala 2076 \u2013 ICEx\/UFMG<\/strong><\/p>

Marcos Oliveira Prates (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Latent archetypes of the spatial patterns of cancer.<\/p>

Resumo:<\/strong> The cancer atlas edited by several countries is the main resource for the analysis of the geographic variation of cancer risk. Correlating the observed spatial patterns with known or hypothesized risk factors is time-consuming work for epidemiologists who need to deal with each cancer separately, breaking down the patterns according to sex and race. The recent literature has proposed to study more than one cancer simultaneously looking for common spatial risk factors. However, this previous work has two constraints: they consider only a very small (2\u20134) number of cancers previously known to share risk factors. In this article, we propose an exploratory method to search for latent spatial risk factors of a large number of supposedly unrelated cancers. The method is based on the singular value decomposition and nonnegative matrix factorization, it is computationally efficient, scaling easily with the number of regions and cancers. We carried out a simulation study to evaluate the method’s performance and apply it to cancer atlas from the USA, England, France, Australia, Spain, and Brazil. We conclude that with very few latent maps, which can represent a reduction of up to 90% of atlas maps, most of the spatial variability is conserved. By concentrating on the epidemiological analysis of these few latent maps a substantial amount of work is saved and, at the same time, high-level explanations affecting many cancers simultaneously can be reached. Joint work with: Thais P. Menezes, Renato M. Assun\u00e7\u00e3o and M\u00f4nica S. M. Castro.<\/p>

10\/01\/2025 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

S\u00e9rgio Felipe Abreu de Britto Bastos (Doutorando, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Infer\u00eancia em Grafos Aleat\u00f3rios atrav\u00e9s de densidades espectrais.<\/p>

Resumo:\u00a0<\/strong>Um grafo consiste em um conjunto de v\u00e9rtices e arestas conectados. As conex\u00f5es podem ser determin\u00edsticas ou aleat\u00f3rias. Neste \u00faltimo, uma possibilidade \u00e9 colocar uma vari\u00e1vel aleat\u00f3ria para determinar se existe ou n\u00e3o uma aresta dado dois v\u00e9rtices, por exemplo. Problemas associados a redes sociais, sinapses neuronais, infec\u00e7\u00e3o de uma doen\u00e7a em uma determinada popula\u00e7\u00e3o, controle de tr\u00e1fego a\u00e9reo, propaga\u00e7\u00e3o de inc\u00eandio em uma floresta, s\u00e3o alguns exemplos onde os grafos aleat\u00f3rios podem ser utilizados para modelar o fen\u00f4meno e fazer infer\u00eancia. A matriz de adjac\u00eancias de um grafo aleat\u00f3rio \u00e9 sim\u00e9trica e seus autovalores s\u00e3o vari\u00e1veis aleat\u00f3rias. A distribui\u00e7\u00e3o dos autovalores associados a uma matriz de adjac\u00eancias de um grafo aleat\u00f3rio \u00e9 chamada de distribui\u00e7\u00e3o espectral que ser\u00e1 caracterizada pela fun\u00e7\u00e3o Delta de Dirac.\u00a0Apresentaremos o resultado de tr\u00eas artigos: (i) Takahashi et al. 2012, (ii) Fujita et al. 2021, e (iii) Preciado et al. 2017.<\/p>

ANO<\/b> DE 2024 – 2\u00ba SEMESTRE<\/strong><\/p>

08\/11\/2024 \u00e0s 13:30h – Local: Zoom e Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Gareth Roberts (University of Warwick, Reino Unido)<\/strong><\/p>

T\u00edtulo:<\/strong> Football group draw probabilities and corrections.<\/p>

Resumo:<\/strong> This talk will consider the challenge of designing football group draw mechanisms which have the uniform distribution over all valid draw assignments, but are also entertaining, practical, and transparent. Although this problem is trivial in completely symmetric problems, it becomes challenging when there are draw constraints which are not exchangeable across each of the competing teams, so that symmetry breaks down. The talk will explain how to simulate the FIFA Sequential Draw method, to compute the non-uniformity of its draws by comparison to a uniform Rejection Sampler. It will then propose two practical methods of achieving the uniform distribution while still using balls and bowls in a way which is suitable for a televised draw. The solutions can also be carried out interactively. The general methodology provided can readily be transported to different competition draws and is not restricted to football events. This is joint work with Jeff Rosenthal.<\/p>

29\/10\/2024 \u00e0s 13:00h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Denis Rustand (KAUST, Ar\u00e1bia Saudita).<\/strong><\/p>

T\u00edtulo:<\/strong> Fast, accurate, and flexible Bayesian survival modeling with the R package INLAjoint.<\/p>

Resumo:<\/strong> This presentation introduces INLAjoint, a user-friendly R package that simplifies the fitting of various survival models using the computationally efficient Integrated Nested Laplace Approximations (INLA) method. INLA offers a significant speed advantage over traditional Markov Chain Monte Carlo (MCMC) methods while maintaining accuracy in parameter estimation. INLAjoint supports a wide range of survival models, including proportional hazards, multi-state, and joint models for multivariate longitudinal and survival data. Joint models, which link multiple regression submodels through correlated or shared random effects, can be computationally intensive. In this context, we underscore the significant reduction in computation time achieved by INLA when compared to MCMC, without compromising on accuracy. Beyond model fitting, the talk provides practical guidance on using the INLAjoint R package, including detailed syntax examples. A key application of joint models is dynamic prediction, which involves estimating the risk of an event (e.g., death or disease progression) based on changes in longitudinal outcomes over time. INLAjoint enables the estimation of dynamic risk predictions and can incorporate updates to these predictions as new longitudinal data becomes available. This makes INLAjoint a valuable tool for analyzing complex health data.<\/p>

25\/10\/2024 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Heitor Soares Ramos Filho (Departamento de Ci\u00eancia da Computa\u00e7\u00e3o, UFMG).<\/strong><\/p>

T\u00edtulo:<\/strong> Deep metric learning e aplica\u00e7\u00f5es.<\/p>

Resumo:<\/strong> A palestra ir\u00e1 abordar o conceito de Deep Metric Learning (DML), uma sub\u00e1rea do aprendizado de m\u00e1quina que se concentra em aprender representa\u00e7\u00f5es de dados em um espa\u00e7o m\u00e9trico. O principal objetivo \u00e9 otimizar a similaridade entre inst\u00e2ncias semelhantes e maximizar a dist\u00e2ncia entre inst\u00e2ncias distintas, utilizando redes neurais profundas. Nessa palestra, iremos introduzir alguns conceitos b\u00e1sicos e aplica\u00e7\u00f5es, enfatizando algumas contribui\u00e7\u00f5es do nosso grupo de pesquisa para a \u00e1rea.<\/p>

11\/10\/2024 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Vin\u00edcius Diniz Mayrink (Departamento de Estat\u00edstica, UFMG).<\/strong><\/p>

T\u00edtulo:<\/strong> Spatial functional data analysis: irregular spacing and Bernstein polynomials.<\/p>

Resumo:<\/strong> Spatial Functional Data (SFD) analysis is an emerging statistical framework that combines Functional Data Analysis (FDA) and spatial dependency modeling. Unlike traditional statistical methods, which treat data as scalar values or vectors, SFD considers data as continuous functions, allowing for a more comprehensive understanding of their behavior and variability. This approach is well-suited for analyzing data collected over time, space, or any other continuous domain. SFD has found applications in various fields, including economics, finance, medicine, environmental science, and engineering. This study proposes new functional Gaussian models incorporating spatial dependence structures, focusing on irregularly spaced data and reflecting spatially correlated curves. The model is based on Bernstein polynomial (BP) basis functions and utilizes a Bayesian approach for estimating unknown quantities and parameters. The study explores the advantages and limitations of the BP model in capturing complex shapes and patterns while ensuring numerical stability. The main contributions of this work include the development of an innovative model designed for SFD using BP, the presence of a random effect to address associations between irregularly spaced observations, and a comprehensive simulation study to evaluate models\u2019 performance under various scenarios. The work also presents one real application of Temperature in Mexico City, showcasing practical illustrations of the proposed model. This is a joint work with Alexander Burbano-Moreno.<\/p>

04\/10\/2024 \u00e0s 13:30h – Local: Zoom e Canal Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Renata Rojas Guerra (Departamento de Estat\u00edstica, UFSM).<\/strong><\/p>

T\u00edtulo:<\/strong> Modelo Rayleigh de escore autorregressivo generalizado para interpreta\u00e7\u00e3o de dados de imagens SAR.<\/p>

Resumo:<\/strong> Este trabalho introduz o modelo Rayleigh de escore autorregressivo generalizado (Ray-GAS), um modelo din\u00e2mico \u00fatil para a interpreta\u00e7\u00e3o de dados de radar de abertura sint\u00e9tica (SAR). Ele \u00e9 derivado da estrutura de escore autorregressivo generalizado (GAS), assumindo que a m\u00e9dia condicional da distribui\u00e7\u00e3o Rayleigh \u00e9 um par\u00e2metro que varia conforme o \u00edndice da imagem. S\u00e3o desenvolvidas ferramentas de estima\u00e7\u00e3o, diagn\u00f3stico e previs\u00e3o para o novo modelo. Al\u00e9m disso, realizamos experimentos num\u00e9ricos com dados simulados de amplitude de uma imagem SAR single-look para dados de regi\u00f5es de floresta e lago. Os resultados ilustram a utilidade do modelo Ray-GAS para a compreens\u00e3o do comportamento estoc\u00e1stico e para a filtragem de retornos de amplitude SAR. Este \u00e9 um trabalho conjunto com Miguel R. Pena-Ramirez e F\u00e1bio M. Bayer.<\/p>

ANO<\/b> DE 2024 – 1\u00ba SEMESTRE<\/strong><\/p>

21\/06\/2024 \u00e0s 13:30h – Local: Zoom e Canal Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

James Sweeney (University of Limerick, Irlanda).<\/strong><\/p>

T\u00edtulo:<\/strong> What is the impact of postcodes on Dublin house prices?<\/p>

Resumo:<\/strong> Accurate and efficient valuation of property is of utmost importance in a variety of settings, including when securing mortgage finance to purchase a property, or where residential property taxes are set as a percentage of a property\u2019s resale value. Internationally, resale based property taxes are most common due to ease of implementation and the difficulty of establishing site values. In an Irish context, property valuations are currently based on comparison to recently sold neighbouring properties. However, this approach is limited by low property turnover. National property taxes based on property value, as opposed to site value, also act as a disincentive to undertake improvement works due to the ensuing increased tax burden. We have developed a spatial hedonic regression model that separates the spatial and non-spatial contributions of property features to resale value. We mitigate the issue of low property turnover through geographic correlation, borrowing information across multiple property types and finishes. We investigate the impact of address mislabelling on predictive performance, where vendors erroneously have given a more affluent postcode, and evaluate the contribution of improvement works to increased values. Our flexible geo-spatial model outperforms all competitors across a number of different evaluation metrics, including the accuracy of both price prediction and associated uncertainty intervals. While our models are applied in an Irish context, the ability to accurately value properties in markets with low property turnover and to quantify the value contributions of specific property features has widespread application. The ability to separate spatial and non-spatial contributions to a property\u2019s value also provides an avenue to site-value based property taxes.<\/p>

14\/06\/2024 \u00e0s 13:30h \u2013 Local: Canal Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Aritra Halder (Drexel University, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Bayesian Modeling with Spatial Curvature Processes<\/p>

Resumo:<\/strong> Spatial process models are widely used for modeling point-referenced variables arising from diverse scientific domains. Analyzing the resulting random surface provides deeper insights into the nature of latent dependence within the studied response. We develop Bayesian modeling and inference for rapid changes on the response surface to assess directional curvature along a given trajectory. Such trajectories or curves of rapid change, often referred to as wombling boundaries, occur in geographic space in the form of rivers in a flood plain, roads, mountains or plateaus or other topographic features leading to high gradients on the response surface. We demonstrate fully model based Bayesian inference on directional curvature processes to analyze differential behavior in responses along wombling boundaries. We illustrate our methodology with a number of simulated experiments followed by multiple applications featuring the Boston Housing data; Meuse river data; and temperature data from the Northeastern United States. Supplementary materials for this article are available online.<\/p>

07\/06\/2024 \u00e0s 13:30h \u2013 Local: Zoom e Canal Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Raquel Borges (Intel Corporation, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Generalized predictive comparisons for complex model interpretation.<\/p>

Resumo:<\/strong> Machine learning algorithms and models constitute the dominant set of predictive methods for a wide range of complex, real-world processes and domains. However, in general, it is difficult to interpret and validate the patterns and insights inferred by the models. We propose a methodology based on generalized predictive comparisons to interpret multiple inputs and interesting functional forms of them to learn and interpret underlying relationships between inputs and the outcome that are inferred by complex models. We demonstrate the broad scope and significance of our generalized predictive comparison methodology by illustrative simulations and case studies.<\/p>

24\/05\/2024 \u00e0s 13:30h \u2013 Local: Sala 2076 \u2013 ICEx\/UFMG<\/strong><\/p>

Guilherme Moura (Depto. de Economia e Rela\u00e7\u00f5es Internacionais, UFSC).<\/strong><\/p>

T\u00edtulo:<\/strong> Regularized Autoregressive Wishart Stochastic Volatility.<\/p>

Resumo:<\/strong> We introduce an extension to Uhlig’s (1994) Wishart Stochastic Volatility model, designed to regularize its covariance predictions toward a specified prior reference matrix. This regularization ensures the stationarity of the observed process and stabilizes the eigenvalues of the predictions. Our method maintains closed-form sequential updating formulas for filtering, prediction, and likelihood evaluation, facilitating practical implementation. Furthermore, we enhance the variance discounting scheme inherent in such models to accommodate varying forgetting rates over time and across different directions in the vector space of observations via directional forgetting. In an empirical portfolio selection application involving up to 1,000 assets, we demonstrate the potential of our proposed approach. It effectively stabilizes the eigenvalues of covariance matrix predictions and generates portfolios with lower risk compared to several benchmark models.<\/p>

10\/05\/2024 \u00e0s 13:30h \u2013 Local: Zoom e Canal do Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Xia Wang (University of Cincinnati, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Variable selection for zero-inflated Poisson regression model.<\/p>

Resumo:<\/strong> The study implements an efficient algorithm for variable selection in the zero-inflated count regression model based on Polya-Gamma latent variables. This leads to a closed form posterior conditional distribution under a logistic link function in modeling the excessive zeros, which helps overcome the computational disadvantage of the logistic link compared to a probit link. Simulation studies examines the efficacy of the proposed model in selecting important variables as well as how the choice of link functions, between the two commonly used probit and the logit links, influences the variable selection results. The proposed model and its comparison with other methods are also illustrated through the application to a German Healthcare dataset. This is a joint work with Haichao Zhang.<\/p>

03\/05\/2024 \u00e0s 13:30h \u2013 Local: Zoom e Canal do Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Havard Rue (KAUST, Ar\u00e1bia Saudita).<\/strong><\/p>

T\u00edtulo:<\/strong> Cross-validation for dependent data.<\/p>

Resumo:<\/strong> I will discuss our new take on cross-validation (CV) for dependent data. Traditional use of CV, like leave-one-out CV, is justified using independence-like assumptions. With dependent data, then leave-one-out CV make less sense, as we are evaluating interpolation properties rather than prediction properties. We can adapt the CV idea to dependent data, by removing a set of “near-by” data-points (to be defined), before predicting, but the issue is then how to do this in practice, which is less evident for more involved models. I will discuss our approach in the context of Latent Gaussian Models (LGM) where we can automatically can select appropriate groups of data to remove before predicting one data point. I will also discuss some new results about group-CV for log-Gaussian Cox processes. The new group-CV approach is available in the R-INLA package.<\/p>

26\/04\/2024 \u00e0s 13:30h – Local: Zoom e Canal do Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Marcelo Bourguignon Pereira (Departamento de Estat\u00edstica, UFRN).<\/strong><\/p>

T\u00edtulo:<\/strong> The weighted beta regression for modeling bounded data.<\/p>

Resumo:<\/strong> A two-parameter weighted beta distribution is introduced for modeling bounded data, which has many similarities to the beta distribution. We propose a class of regression models where the response is weighted beta distributed and the two shape parameters that index weighted the beta distribution are related to covariates and regression parameters. The proposed regression model is a natural strong competitor of the beta regression model. We study mathematical and statistical properties of the distribution and we provide a useful interpretation of the parameters. The maximum likelihood method is used for estimating the model parameters. Simulation studies are conducted to investigate the performance of the maximum likelihood estimators and the asymptotic confidence intervals of the parameters. An application of the proposed regression model to real bounded data is presented. Trabalho em conjunto com os professores Diego I. Gallardo (Universidad del B\u00edo-B\u00edo) e Roberto Vila (UnB).<\/p>

19\/04\/2024 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Adrian P. H. Luna (Departamento de Estat\u00edstica, UFMG).<\/strong><\/p>

T\u00edtulo:<\/strong> Redes Neurais Gr\u00e1ficas GGN.<\/p>

Resumo:<\/strong> As redes neurais s\u00e3o parte dos m\u00e9todos de intelig\u00eancia artificial (IA) mais populares no solu\u00e7\u00e3o de problemas complexos onde temos muita informa\u00e7\u00e3o dispon\u00edvel. Parte importante destes m\u00e9todos \u00e9 composta pelas redes de convolu\u00e7\u00e3o. Redes neurais gr\u00e1ficas, GGN, s\u00e3o modelos de redes que tem operadores de convolu\u00e7\u00e3o gr\u00e1fica e servem para o caso de ter a informa\u00e7\u00e3o estruturada na forma de grafos, por exemplo informa\u00e7\u00e3o espacial, estruturas sem\u00e2nticas, redes de colabora\u00e7\u00e3o, etc. Resultados recentes sobre o comportamento assint\u00f3tico das redes neurais para a determina\u00e7\u00e3o da velocidade de converg\u00eancia as solu\u00e7\u00f5es das redes usam aproxima\u00e7\u00f5es das redes neurais aos modelos da mec\u00e2nica estat\u00edstica, o que nos permite caracterizar melhor o problema da n\u00e3o convexidade m\u00faltipla associado as redes neurais, e como se reflete isso nas redes GGN. Vamos apresentar tamb\u00e9m uma aplica\u00e7\u00e3o das redes neurais, usando um modelo de GGN para a previs\u00e3o epidemiol\u00f3gica da dengue no campus da UFMG.<\/p>

12\/04\/2024 \u00e0s 13:30h \u2013 Local: Zoom e Canal do Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Matthias Katsfuss (University of Wisconsin-Madison, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Probabilistic function estimation via nearest-neighbor directed acyclic graphs.<\/p>

Resumo:<\/strong> We consider probabilistic inference on continuous functions or fields, such as time series, geospatial fields, response surfaces of computer models, or regression functions. Gaussian processes (GPs) are popular models for such applications, but Gaussian assumptions are too restrictive in many settings. Sparse autoregressive structures corresponding to nearest-neighbor directed acyclic graphs (NN-DAGs) can lead to scalable, accurate, and flexible inference. We provide a number of examples, including so-called Vecchia approximations of GPs, and autoregressive GPs for learning high-dimensional spatial distributions from a small number of training samples (e.g., for climate-model emulation). When the function of interest is latent, we propose a novel framework for variational inference targeting its potentially non-Gaussian posterior. We make NN-DAG assumptions for both the prior and variational families, with highly expressive conditional distributions in the variational family. Scalable model fitting can be achieved via doubly stochastic variational optimization with polylogarithmic time complexity per iteration based on reduced ancestor sets.<\/p>

05\/04\/2024 \u00e0s 13:30h \u2013 Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Uriel M. Silva (Departamento de Estat\u00edstica, UFMG).<\/strong><\/p>

T\u00edtulo:<\/strong> T\u00f3picos em Infer\u00eancia para Modelos de Espa\u00e7o de Estados via SMC.<\/p>

Resumo:<\/strong> Neste semin\u00e1rio ser\u00e1 apresentada a teoria b\u00e1sica de infer\u00eancia via SMC (Sequential Monte Carlo, tamb\u00e9m conhecidos como Filtros de Part\u00edculas) em modelos de Espa\u00e7o de Estados, assim como alguns problemas em aberto na \u00e1rea. Ser\u00e3o discutidos aspectos de infer\u00eancia Cl\u00e1ssica e Bayesiana para os par\u00e2metros dessa classe de modelos, e em particular a possibilidade de implementa\u00e7\u00e3o de algoritmos do tipo RMHMC (Riemannian Manifold Hamiltonian Monte Carlo).<\/p>

22\/03\/2024 \u00e0s 13:30h – Local: Zoom e Canal do Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Ying MacNab (University of British Columbia, Canada).<\/strong><\/p>

T\u00edtulo:<\/strong> On Gaussian Markov random field, spatial dependence representation, and local influence function.<\/p>

Resumo:<\/strong> Gaussian Markov random fields (GMRF) and their multivariate extensions (MGMRFs) are powerful tools for modeling probabilistic interactions of directly related variables. As an important category of graphical models, they are commonly used in spatial statistics (e.g., disease mapping, small area estimation, spatial ecology) and Bayesian statistics, and their applications and potentials of application are far-reaching (e.g., artificial intelligence, deep learning, image processing, computer vision, spatial biology). In this presentation, I give an overview of my recent work on spatial dependence representations for selected (adaptive) (M)GMRF parameterizations and introduce the notions of (a)symmetric local influence, cross-local influence, and associated local and cross-local influence functions. Some recent applications in the contexts of Bayesian (spatial, multivariate, and spatiotemporal) disease mapping and small-area estimation will be presented.<\/p>

15\/03\/2024 \u00e0s 13:30h – Local: Zoom e Canal do Youtube Semin\u00e1rios DEST-UFMG<\/strong><\/p>

Pedro Luiz Ramos (PUC, Chile)<\/strong><\/p>

T\u00edtulo:<\/strong> Asymptotic properties of generalized closed-form maximum likelihood estimators.<\/p>

Resumo:<\/strong> The maximum likelihood estimator (MLE) is pivotal in statistical inference, yet its application is often hindered by the absence of closed-form solutions for many models. This poses challenges in real-time computation scenarios, particularly within embedded systems technology, where numerical methods are impractical. This study introduces a generalized form of the MLE that yields closed-form estimators under certain conditions. We derive the asymptotic properties of the proposed estimator and demonstrate that our approach retains key properties such as invariance under one-to-one transformations, strong consistency, and an asymptotic normal distribution. The effectiveness of the generalized MLE is exemplified through its application to the Gamma, Nakagami, and Beta distributions, showcasing improvements over the traditional MLE. Additionally, we extend this methodology to a bivariate gamma distribution, successfully deriving closed-form estimators. This advancement presents significant implications for real-time statistical analysis across various applications.<\/p>

ANO<\/b> DE 2023 – 2\u00ba SEMESTRE<\/strong><\/p>

24\/11\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Samuel Faria C\u00e2ndido (Doutorando, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Bayesian Nonstationary and Nonparametric Covariance Estimation for Large Spatial Data.<\/p>

Resumo:<\/strong> In spatial statistics, it is often assumed that the spatial field of interest is stationary and its covariance has a simple parametric form, but these assumptions are not appropriate in many applications. Given replicate observations of a Gaussian spatial field, we propose nonstationary and nonparametric Bayesian inference on the spatial dependence. Instead of estimating the quadratic (in the number of spatial locations) entries of the covariance matrix, the idea is to infer a near-linear number of nonzero entries in a sparse Cholesky factor of the precision matrix. Our prior assumptions are motivated by recent results on the exponential decay of the entries of this Cholesky factor for Matern-type covariances under a specific ordering scheme. Our methods are highly scalable and parallelizable. We conduct numerical comparisons and apply our methodology to climate-model output, enabling statistical emulation of an expensive physical model. Reference: Kidd B. and Katzfuss M. (2022) Bayesian Analysis, 17, 1, 291-351.<\/p>

17\/11\/2023 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Ma\u00edra Soalheiro (Doutoranda, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Modelling for Poisson process intensities over irregular spatial domains.<\/p>

Resumo:<\/strong> We develop nonparametric Bayesian modelling approaches for Poisson processes, using weighted combinations of structured beta densities to represent the point process intensity function. For a regular spatial domain, such as the unit square, the model construction implies a Bernstein-Dirichlet prior for the Poisson process density, which supports general inference for point process functionals. The key contribution of the methodology is two classes of flexible and computationally efficient models for spatial Poisson process intensities over irregular domains. We address the choice or estimation of the number of beta basis densities and develop methods for prior specification and posterior simulation for full inference about functionals of the point process. The methodology is illustrated with both synthetic and real data sets. Reference: Zhao C. and Kottas A. (2021) Preprint arXiv:2106.04654<\/p>

10\/11\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Abhirup Datta (Department of Biostatistics, Johns Hopkins University, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong> Combining machine learning with Gaussian processes for geospatial data.<\/p>

Resumo:<\/strong> Spatial generalized linear mixed models, consisting of a linear covariate effect and a Gaussian Process (GP) distributed spatial random effect, are widely used for analyses of geospatial data. We consider the setting where the covariate effect is non-linear and propose modeling it using a flexible machine learning algorithm like random forests or deep neural networks. We propose well-principled extensions of these methods, for estimating non-linear covariate effects in spatial mixed models where the spatial correlation is still modeled using GP. The basic principle is guided by how ordinary least squares extends to generalized least squares for linear models to account for dependence. We demonstrate how the same extension can be done for these machine learning approaches like random forests and neural networks. We provide extensive theoretical and empirical support for the methods and show how they fare better than na\u00efve or brute-force approaches to use machine learning algorithms for spatially correlated data. We demonstrate the RandomForestsGLS R-package that implements this extension for random forests.<\/p>

20\/10\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Fernando A. Quintana (PUC, Chile).<\/strong><\/p>

T\u00edtulo:<\/strong> Childhood Obesity in Singapore: a Bayesian Nonparametric Approach<\/p>

Resumo:<\/strong> Overweight and obesity in adults are known to be associated with increased risk of metabolic and cardiovascular diseases. Obesity has now reached epidemic proportions, increasingly affecting children. Therefore, it is important to understand if this condition persists from early life to childhood and if different patterns can be detected to inform intervention policies. Our motivating application is a study of temporal patterns of obesity in children from South Eastern Asia. Our main focus is on clustering obesity patterns after adjusting for the effect of baseline information. Specifically, we consider a joint model for height and weight over time. Measurements are taken every six months from birth. To allow for data-driven clustering of trajectories, we assume a vector autoregressive sampling model with a dependent logit stick-breaking prior. Simulation studies show good performance of the proposed model to capture overall growth patterns, as compared to other alternatives. We also fit the model to the motivating dataset, and discuss the results, in particular highlighting cluster differences. We have found four large clusters, corresponding to children sub-groups, though two of them are similar in terms of both height and weight at each time point. We provide an interpretation of these clusters in terms of combinations of predictors.<\/p>

29\/09\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Luis Carvalho (Boston University, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong> Daviance matrix factorization.<\/p>

Resumo:<\/strong> We investigate a general matrix factorization for deviance-based data losses, extending the ubiquitous singular value decomposition beyond squared error loss. While similar approaches have been explored before, our method leverages classical statistical methodology from generalized linear models (GLMs) and provides an efficient algorithm that is flexible enough to allow for structural zeros via entry weights. Moreover, by adapting results from GLM theory, we provide support for these decompositions by (i) showing strong consistency under the GLM setup, (ii) checking the adequacy of a chosen exponential family via a generalized Hosmer-Lemeshow test, and (iii) determining the rank of the decomposition via a maximum eigenvalue gap method. To further support our findings, we conduct simulation studies to assess robustness to decomposition assumptions and extensive case studies using benchmark datasets from image face recognition, natural language processing, network analysis, and biomedical studies. Our theoretical and empirical results indicate that the proposed decomposition is more flexible, general, and robust, and can thus provide improved performance when compared to similar methods. To facilitate applications, an R package with efficient model fitting and family and rank determination is also provided.<\/p>

22\/09\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Fernanda L. Schumacher (The Ohio State University, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong> Penalized Estimation of Scale Mixture Of Skew-Normal Linear Mixed Models using Hamiltonian Monte Carlo<\/p>

Resumo:<\/strong> In clinical trials, studies often present longitudinal or clustered data. These studies are commonly analyzed using linear mixed models, and for mathematical convenience, it is usually assumed that both random effect and error term follow normal distributions. These restrictive assumptions, however, may result in a lack of robustness against departures from the normal distribution and invalid statistical inferences. An interesting extension to make these models more flexible by accounting for skewness and heavy tails is considering the scale mixture of skew-normal class of distributions. Nevertheless, a practical problem may arise when modeling distributions derived from the skew-normal: the possibility that the maximum likelihood estimate of the parameter that regulates skewness diverges. In this work, this anomaly is illustrated, and an alternative Bayesian estimation via Hamiltonian Monte Carlo is proposed.<\/p>

15\/09\/2023 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Wilhelm Alexander C. Steinmetz (Departamento de Matem\u00e1tica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Fundamentos da Matem\u00e1tica e Filosofia Empiricamente Informada.<\/p>

Resumo:<\/strong> Nesta palestra, procuro abordar como outras ci\u00eancias como Ci\u00eancia Cognitiva, Neuroci\u00eancia, Antropologia Cultural, Psicologia do Desenvolvimento, Pedagogia da Matem\u00e1tica e Hist\u00f3ria da Matem\u00e1tica podem lan\u00e7ar um luz sobre quest\u00f5es filos\u00f3ficas referentes aos Fundamentos da Matem\u00e1tica.<\/p>

01\/09\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Gregory J. Matthews (University of Loyola Chicago, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Completion of Partially Observed Curves Using Hot Deck Type Imputation<\/p>

Resumo:<\/strong> Statistical shape analysis of curves is well-developed when curves are fully observed. This work considers partially observed curves and develops methods for curve completion or imputation by leveraging tools from the statistical analysis of shape of fully observed curves, which enables sensible curve completions. On a dataset containing partially observed bovid teeth arising from a biological anthropology application, the method is implemented and classification of the completed teeth is carried out based on a shape distance on the set of curves.<\/p>

18\/08\/2023 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Shariq Mohammed (Departamento de Bioestat\u00edstica, Boston University, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong> Layered Variable Selection for Multivariate Bayesian Regression: A Case Study in Imaging-Genomics<\/p>

Resumo:<\/strong> We propose a statistical framework to integrate radiological magnetic resonance imaging (MRI) and genomic data to identify the underlying radiogenomic associations in lower-grade gliomas (LGG). We devise a novel imaging phenotype by dividing the tumor region into concentric spherical layers that mimic the tumor evolution process. MRI data within each layer is represented by voxel-intensity-based probability density functions which capture the complete information about tumor heterogeneity. Under a Riemannian-geometric framework, these densities are mapped to a vector of principal component scores which act as imaging phenotypes. Subsequently, we build Bayesian variable selection models for each layer with the imaging phenotypes as the response and the genomic markers as predictors. Our novel hierarchical prior formulation incorporates the interior-to-exterior structure of the layers and the correlation between the genomic markers. We employ a computationally efficient Expectation-Maximization-based strategy for estimation. With a focus on the cancer driver genes in LGG, we discuss some biologically relevant findings.<\/p>

ANO<\/b> DE 2023 – 1\u00ba SEMESTRE<\/strong><\/p>

30\/06\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Adriana dos Santos Lima (Doutoranda, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Regression analysis of interval-censored failure time data with possibly crossing hazards<\/p>

Resumo:<\/strong> Interval-censored failure time data occur in many areas, especially in medical follow-up studies such as clinical trials, and in consequence, many methods have been developed for the problem. However, most of the existing approaches cannot deal with situations where the hazard functions may cross each other. To address this, we develop a sieve maximum likelihood estimation procedure with the application of the short-term and long-term hazard ratio model. In the method, the I-splines are used to approximate the underlying unknown function. \/n\/ extensive simulation study was conducted for the assessment of the finite sample properties of the presented procedure and suggests that the method seems to work well for practical situations. The analysis of a motivated example is also provided. Reference: Zhang H., Wang P. and Sun J. (2018) Statistics in Medicine, 37, 5, 786-775.<\/p>

23\/06\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Michael Willig (University of Connecticut, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong> Patterns in ecology: Circular statistics, methodological concerns and bootstrap approaches.<\/p>

Resumo:<\/strong> In this talk, we present results of an ongoing project designed to (1) demonstrate, via a number of exemplar data sets, how application of classical circular statistics in some designs can lead to erroneous and counterintuitive conclusions; (2) develop a bootstrap approach to overcome limitations associated with marginal totals; (3) apply this bootstrap approach to the exemplar data sets to highlight its salient improvement; and (4) apply both circular statistics (i.e., Rayleigh and Hermans-Rasson Tests) and the proposed boot-strap approach to reproductive phenologies derived from well-studied mammal species from the Amazon of Peru. Finally, we wish to promote collaborations between statisticians and ecologists to address questions in temporal biology.<\/p>

02\/06\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG <\/strong><\/p>

Clarice G. B. Dem\u00e9trio (ESALQ – USP)<\/strong><\/p>

T\u00edtulo:<\/strong> Extended Poisson-Tweedie: properties and regression models for count data.<\/p>

Resumo:<\/strong> We propose a new class of discrete generalized linear models based on the class of Poisson-Tweedie factorial dispersion models with variance of the form, where $ is the mean, and p are the dispersion and Tweedie power parameters, respectively (Bonat et al, 2018; 18: 24\u201349). The models are fitted by using an estimating function approach obtained by combining the quasi-score and Pearson estimating functions for estimation of the regression and dispersion parameters, respectively. This provides a flexible and efficient regression methodology for a comprehensive family of count models including Hermite, Neyman Type A, P\u00f3lya-Aeppli, negative binomial and Poisson-inverse Gaussian. The estimating function approach allows us to extend the Poisson-Tweedie distributions to deal with underdispersed count data by allowing negative values for the dispersion parameter. Furthermore, the Poisson-Tweedie family can automatically adapt to highly skewed count data with excessive zeros, without the need to introduce zero-inflated or hurdle components, by the simple estimation of the power parameter. Thus, the proposed models offer a unified framework to deal with under, equi, overdispersed, zero-inflated and heavy-tailed count data. The computational implementation of the proposed models is fast, relying only on a simple Newton scoring algorithm. Simulation studies showed that the estimating function approach provides unbiased and consistent estimators for both regression and dispersion parameters. We highlight the ability of the Poisson-Tweedie distributions to deal with count data through a consideration of dispersion, zero-inflated and heavy tail indexes, and illustrate its application with four data analyses.<\/p>

26\/05\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Ang\u00e9lica M. Tortola Ribeiro (Universidade Tecnol\u00f3gica Federal do Paran\u00e1)<\/strong><\/p>

T\u00edtulo:<\/strong> A Kronecker-based covariance model for multivariate geostatistical data<\/p>

Resumo:<\/strong> In this work, we present a proposal for a covariance function specification for spatially continuous multivariate data. This model is based on the Kronecker product of covariance matrices for Gaussian random fields. The structure is valid for different marginal covariance functions, allowing different variables to have different spatial dependence structures, which makes it more flexible. Our model allows its parameters to vary in its usual domains, which makes the estimation less constrained when compared to other classical approaches. The reduced computational times and easy generalization to larger dimensions follows from the model definition. The simple structure of the model, combined with the interpretability of the parameters and computational time for inference make this model a promising candidate for modeling spatially continuous multivariate data.<\/p>

19\/05\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Tamara Broderick (Department of Electrical Engineering and Computer Science, MIT, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> An Automatic Finite-Sample Robustness Check: Can Dropping a Little Data Change Conclusions?<\/p>

Resumo:<\/strong> Practitioners will often analyze a data sample with the goal of applying any conclusions to a new population. For instance, if economists conclude microcredit is effective at alleviating poverty based on observed data, policymakers might decide to distribute microcredit in other locations or future years. Typically, the original data is not a perfect random sample from the population where policy is applied — but researchers might feel comfortable generalizing anyway so long as deviations from random sampling are small, and the corresponding impact on conclusions is small as well. Conversely, researchers might worry if a very small proportion of the data sample was instrumental to the original conclusion. So we propose a method to assess the sensitivity of statistical conclusions to the removal of a very small fraction of the data set. Manually checking all small data subsets is computationally infeasible, so we propose an approximation based on the classical influence function. Our method is automatically computable for common estimators. We provide finite-sample error bounds on approximation performance and a low-cost exact lower bound on sensitivity. We find that sensitivity is driven by a signal-to-noise ratio in the inference problem, does not disappear asymptotically, and is not decided by misspecification. Empirically we find that many data analyses are robust, but the conclusions of several influential economics papers can be changed by removing (much) less than 1% of the data.<\/p>

12\/05\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Alexandre L. Rodrigues (Departamento de Estat\u00edstica, UFES).<\/strong><\/p>

T\u00edtulo:<\/strong> A conditional machine learning classification approach for spatio-temporal risk assessment of crime data.<\/p>

Resumo:<\/strong> Crime data analysis is an essential source of information to aid social and political decisions makers regarding the allocation of public security resources. Computer-aided dispatch systems and technological advances in geographic information systems have made analysing and visualising historical spatial and temporal records of crimes a vital part of police operations and strategy. We look at our motivating crime problem as a spatio-temporal point pattern. Using a conditional approach based on properties of Poisson point processes, we transform the spatio-temporal point process prediction problem into a classification problem. We create spatio-temporal handcrafted features to link future and past events and use machine learning algorithms to learn behavioural patterns from the data. The fitted model is then used to carry out the reverse transformation, i.e. to perform spatio-temporal risk predictions based on the outcomes of the classification problem. Our procedure has theoretical formalism from point process theory and gains flexibility and computational efficiency inherited from the machine learning field. We show its performance under some simulated scenarios and a real application to spatio-temporal prediction and risk assessment of homicides in Bogota, Colombia.<\/p>

05\/05\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Vald\u00e9rio A. Reisen (UFES)<\/strong><\/p>

T\u00edtulo:<\/strong> M-quatile estimation for GARCH models.<\/p>

Resumo:<\/strong> M-regression and quantile methods have been suggested to estimate generalized autoregressive conditionally heteroscedastic (GARCH) models. In this paper, we propose an M-quantile approach, which combines quantile and M-regression to obtain a robust estimator of the conditional volatility when the data have abrupt observations or heavy-tailed distributions. Some technical issues are discussed and Monte Carlo experiments are conducted to show that the M-quantile approach appears to be more resistant against additive outliers than M-regression and quantile methods. The usefulness of the method is illustrated on two financial datasets.<\/p>

28\/04\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Bruno Sans\u00f3 (University of California Santa Cruz, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Non-Gaussian geostatistical models using nearest neighbors processes.<\/p>

Resumo:<\/strong> We present a framework for non-Gaussian spatial processes that encompasses large distribution families. Spatial dependence for a set of irregularly scattered locations is described with a mixture of pairwise kernels. Focusing on the nearest neighbors of a given location, within a reference set, we obtain a valid spatial process: the nearest neighbor mixture process (NNMP). We develop conditions to construct general NNMP models with arbitrary pre-specified marginal distributions. Essentially, NNMPs are specified by a bi-variate distribution, with suitable marginals, used to specify the mixture transition kernels. Such distribution can be spatially varying, to capture non-homogeneous spatial features. The mixture structure of the model allows for efficient MCMC-based exploration of posterior distribution of the model parameters, even for relatively large number of locations. We illustrate the capabilities of NNMPs with observations corresponding to distributions with different non-Gaussian characteristics: Long tails; Compact support; Skewness; Discrete values.<\/p>

14\/04\/2023 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Gabriel O. Assun\u00e7\u00e3o (IBGE)<\/strong><\/p>

T\u00edtulo:<\/strong> Aspectos Metodol\u00f3gicos do Sistema Integrado de Pesquisas Domiciliares.<\/p>

Resumo:<\/strong> As pesquisas domiciliares amostrais realizadas pelo Instituto Brasileiro de Geografia e Estat\u00edstica (IBGE) s\u00e3o fundamentais para retratar o Brasil com informa\u00e7\u00f5es necess\u00e1rias ao conhecimento da sua realidade e ao exerc\u00edcio da cidadania, al\u00e9m de serem essenciais para a formula\u00e7\u00e3o de pol\u00edticas p\u00fablicas. Nesse sentido, o Sistema Integrado de Pesquisas Domiciliares (SIPD) foi implementado pelo IBGE em 2011 com o intuito de integrar todas as pesquisas domiciliares amostrais a partir da utiliza\u00e7\u00e3o de uma mesma infraestrutura amostral, de um mesmo cadastro de sele\u00e7\u00e3o e de uma amostra comum, a Amostra Mestra. Ent\u00e3o, o intuito deste semin\u00e1rio \u00e9 apresentar sobre os aspectos metodol\u00f3gicos relacionados ao SIPD e \u00e0 Amostra Mestra.<\/p>

31\/03\/2023 \u00e0s 13:30h – Local: Canal do Youtube Semin\u00e1rios DEST \u2013 UFMG<\/strong><\/p>

Jorge Mateu (Universitat Jaume I, Espanha).<\/strong><\/p>

T\u00edtulo:<\/strong> Statistical models for the analysis, prediction and monitoring of space-time data. Applications to infectious diseases and crime.<\/p>

Resumo:<\/strong> We present several statistical approaches to understand the underlying temporal and spatial dynamics of infectious diseases (with a focus on Covid-19 data) that can result in informed and timely public health policies. Most studies in the context of infectious diseases commonly report figures of the overall infection at a state- or county-level, reporting the aggregated number of cases in a particular region at one time. However, we focus on analysing high-resolution Covid-19 datasets in form of spatio-temporal point patterns, offering vital insights for the spatio-temporal interaction between individuals concerning the disease spread in a metropolis.<\/p>

24\/03\/2023 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Carl Schmertmann (Florida State University, EUA).<\/strong><\/p>

T\u00edtulo:<\/strong> Estima\u00e7\u00e3o bayesiana de mortalidade para pequenas \u00e1reas com sub-registro de \u00f3bitos.<\/p>

Resumo:<\/strong> Vari\u00e2ncia amostral dificulta a estima\u00e7\u00e3o de taxas demogr\u00e1ficas para pequenas \u00e1reas. Al\u00e9m disso, em muitos pa\u00edses o sistema de registro de \u00f3bitos \u00e9 imperfeito, com um grau de cobertura que varia entre regi\u00f5es. Elaboramos um modelo bayesiano para mortalidade que lida com esses dois problemas simultaneamente. O modelo incorpora estimativas externas do sub-registro local atrav\u00e9s de distribui\u00e7\u00f5es a priori para os par\u00e2metros que definam o grau de cobertura. Aplicamos o modelo a dados de 2009-2011 para gerar estimativas de taxas de mortalidade e esperan\u00e7a de vida ao nascer — e da incerteza nessas estimativas — para todas as microrregi\u00f5es brasileiras.<\/p>

17\/03\/2023 \u00e0s 13:30h – Local: Sala 2076 – ICEx\/UFMG<\/strong><\/p>

Bernardo N. B. Lima (Departamento de Matem\u00e1tica, UFMG).<\/strong><\/p>

T\u00edtulo:<\/strong> B\u00eabados, apostas e circuitos el\u00e9tricos.<\/p>

Resumo:<\/strong> O passeio aleat\u00f3rio \u00e9 um dos objetos mais interessantes e estudados em Probabilidade. Descreveremos a mais simples de suas vers\u00f5es e exploraremos, tamb\u00e9m no caso mais simples, uma teoria matem\u00e1tica que tamb\u00e9m \u00e9 comum em outros contextos aparentemente bem distintos.<\/p>

ANO<\/b> DE 2022 – 2\u00ba SEMESTRE<\/strong><\/p>

02\/12\/2022 \u00e0s 13:00 hs\u00a0–\u00a0<\/strong>Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Gracielle A. Ara\u00fajo (Doutoranda, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Bayesian methods for neural networks and related models.<\/p>

Resumo:\u00a0<\/strong>Models such as feed-forward neural networks and certain other structures investigated in the computer science literature are not amenable to closed-form Bayesian analysis. The paper reviews the various approaches taken to overcome this difficulty, involving the use of Gaussian approximations, Markov chain Monte Carlo simulation routines, and a class of non-Gaussian but \u201cdeterministic\u201d approximations called variational approximations. Reference: Titterington D. M. (2004), Bayesian methods for neural networks and related models. Statistical Science, 19, 1, 128-139.<\/p>

25\/11\/2022 \u00e0s 13:30 hs\u00a0–\u00a0<\/strong>Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Caio G. B. Balieiro (Doutorando, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modeling spatial variation in leukemia survival data.<\/p>

Resumo:\u00a0<\/strong>In this article we combine ideas from spatial statistics with lifetime data analysis techniques to investigate possible spatial variation in survival of adult acute myeloid leukemia patients in northwest England. Exploratory analysis suggests both clinically and statistically significant variation in survival rates across the region. A multivariate gamma frailty model incorporating spatial dependence is proposed and applied, with results confirming the dependence of hazard on location. Reference: Henderson R., Shimakura S. and Gorst D. (2002), Modeling spatial variation in leukemia survival data. Journal of the American Statistical Association, 97, 460, 965-972.<\/p>

18\/11\/2022 \u00e0s 13:30 hs\u00a0–\u00a0<\/strong>Local:\u00a0sala 2076 – ICEx\/UFMG<\/strong><\/p>

Marco Antonio T. Aucahuasi (Doutorando, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A brief overview of Markov chains and coalescing particles.<\/p>

Resumo:\u00a0<\/strong>In this talk we present a brief review of the theory of Markov chains and mixing times, and some examples. We also present an interacting particle system with the following dynamics: At time 0, we begin with a particle at each integer in [0,n]. At each positive integer time, one of the particles remaining in [1,n] is chosen at random and moves one to the left, coalescing with any particle that might already be there. How long does it take until all particles coalesce (at 0)? Orientador: Roger W. C. Silva.<\/p>

11\/11\/2022 \u00e0s 13:30 hs\u00a0–\u00a0<\/strong>Local:\u00a0sala 2076 – ICEx\/UFMG<\/strong><\/p>

Ot\u00e1vio A. S. Lima (Doutorando, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0O modelo de percola\u00e7\u00e3o de palavras.<\/p>

Resumo:\u00a0<\/strong>O problema de percola\u00e7\u00e3o de palavras foi introduzido por Itai Benjamini e Harry Kesten em 1995, como generaliza\u00e7\u00e3o do modelo de percola\u00e7\u00e3o Bernoulli. Esta apresenta\u00e7\u00e3o tem como objetivo introduzir este modelo e apresentar alguns resultados.<\/p>

04\/11\/2022 \u00e0s 13:30 hs\u00a0–\u00a0<\/strong>Local:\u00a0sala 2076 – ICEx\/UFMG<\/strong><\/p>

Tha\u00eds P. Galletti (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Gera\u00e7\u00e3o de coordenadas geogr\u00e1ficas sint\u00e9ticas para banco de dados confidenciais com aplica\u00e7\u00e3o a dados de COVID-19 em Montes Claros, MG.<\/p>

Resumo:\u00a0<\/strong>Com a crescente produ\u00e7\u00e3o de dados das \u00faltimas d\u00e9cadas, um dos principais problemas \u00e9 a viola\u00e7\u00e3o da privacidade de indiv\u00edduos. O desafio \u00e9 desenvolver mecanismos que preservem o sigilo dos dados e, ao mesmo tempo, permitam que os dados sejam divulgados e utilizados para an\u00e1lises estat\u00edsticas. Os m\u00e9todos de imputa\u00e7\u00e3o m\u00faltipla para simula\u00e7\u00e3o de dados sint\u00e9ticos t\u00eam se mostrado uma alternativa interessante para resolver esse tipo de problema, podendo ser aplicado inclusive para localiza\u00e7\u00f5es espaciais. O objetivo deste trabalho \u00e9 propor uma extens\u00e3o para a metodologia de gera\u00e7\u00e3o de coordenadas geogr\u00e1ficas sint\u00e9ticas com covari\u00e1veis discretas e cont\u00ednuas, al\u00e9m de aplicar o m\u00e9todo para imputa\u00e7\u00e3o de localiza\u00e7\u00f5es sint\u00e9ticas de indiv\u00edduos com suspeita de COVID-19 na cidade de Montes Claros, MG.<\/p>

21\/10\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Rafael Izbicki (Departamento de Estat\u00edstica, UFSCar)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Diagnostics and recalibration of predictive distributions.<\/p>

Resumo:\u00a0<\/strong>Uncertainty quantification is crucial for assessing the predictive ability of AI algorithms. A large body of work (including normalizing flows and Bayesian neural networks) has been devoted to describing the entire predictive distribution (PD) of a target variable Y given input features X. However, off-the-shelf PDs are usually far from being conditionally calibrated; i.e., the probability of occurrence of an event given input X can be significantly different from the predicted probability. Most current research on predictive inference (such as conformal prediction) concerns constructing calibrated prediction sets only. It is often believed that the problem of obtaining and assessing entire conditionally calibrated PDs is too challenging. In this work, we show that recalibration, as well as validation of full\/entire PDs, are indeed attainable goals in practice. Our proposed method relies on the idea of regressing probability integral transform (PIT) scores against X. This regression gives full diagnostics of conditional coverage across the entire feature space and can be used to recalibrate misspecified PDs. We benchmark our corrected prediction bands against oracle bands and state-of-the-art predictive inference algorithms for synthetic data, including settings with a distributional shift. Finally, we produce calibrated PDs for two applications: (i) probabilistic forecasting based on sequences of satellite images, and (ii) estimation of galaxy distances based on imaging data (photometric redshifts).<\/p>

14\/10\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a>
<\/a><\/strong><\/p>

F\u00e1bio M. Bayer (Departamento de Estat\u00edstica, UFSM)<\/strong><\/p>

T\u00edtulo:<\/strong> K vizinhos mais pr\u00f3ximos circular.<\/p>

Resumo:\u00a0<\/strong>Dados circulares est\u00e3o presentes em v\u00e1rias \u00e1reas da ci\u00eancia e carecem de m\u00e9todos estat\u00edsticos espec\u00edficos para seu tratamento. No \u00e2mbito de modelos de regress\u00e3o, a literatura apresenta modelos de regress\u00e3o param\u00e9tricos para dados circulares, os quais fazem suposi\u00e7\u00f5es de determinadas distribui\u00e7\u00f5es de probabilidade circulares para seus ajustes. Por outro lado, na \u00e1rea de aprendizado de m\u00e1quina, uma abordagem supervisionada para predi\u00e7\u00e3o de dados cont\u00ednuos envolve modelos de regress\u00e3o n\u00e3o param\u00e9tricos, os quais podem n\u00e3o ser adequados para situa\u00e7\u00f5es em que a vari\u00e1vel resposta \u00e9 circular. Neste semin\u00e1rio, apresentarei um novo modelo de aprendizado de m\u00e1quina para predi\u00e7\u00e3o de dados circulares, o qual \u00e9 denominado k vizinhos mais pr\u00f3ximos circular. Trabalho co-autorado com Maicon Facco.<\/p>

07\/10\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a>
<\/a><\/strong><\/p>

Vald\u00e9rio A. Reisen (UFES, UFMG, Universit\u00e9 Paris-Saclay, UFBA)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0M-regression estimation methods and robust PCA in mixed linear models. An application to quantify the statistical association between forced expiratory volume and pollutants.<\/p>

Resumo:\u00a0<\/strong>This seminar discusses the use of M-regression estimation methods and PCA tools (robust and non-robust) in Mixed models with time series covariates. An application to the relationship between exposure to air pollution and forced expiratory volume at the first second (FEV1) is considered to motivate the use of the proposed methodology in real problems.<\/p>

30\/09\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0sala 2076 – ICEx\/UFMG
<\/a><\/strong><\/p>

Jussiane N. Gon\u00e7alves (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A novel regression model for correlated count data.<\/p>

Resumo:\u00a0<\/strong>The premise of independence among subjects in the same cluster\/group often fails in practice, and models that rely on such untenable assumption can produce misleading results. To overcome this severe deficiency, we introduce a new regression model to handle overdispersed and correlated clustered counts. To account for correlation within clusters, we propose a Poisson regression model where the observations within the same cluster are driven by the same latent random effect that follows the Birnbaum-Saunders distribution with a parameter that controls the strength of dependence among the individuals. This novel multivariate count model is called Clustered Poisson Birnbaum-Saunders (CPBS) regression. The CPBS model is analytically tractable, and its moment structure can be explicitly obtained. Estimation of parameters is performed through the maximum likelihood method, and an Expectation-Maximization (EM) algorithm is also developed. Simulation results to evaluate the finite-sample performance of our proposed estimators are presented. We also discuss diagnostic tools for checking model adequacy. An empirical application concerning the number of inpatient admissions by individuals to hospital emergency rooms, from the Medical Expenditure Panel Survey (MEPS) conducted by the United States Agency for Health Research and Quality, illustrates the usefulness of our proposed methodology..<\/p>

23\/09\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0sala 2076 – ICEx\/UFMG
<\/a><\/strong><\/p>

F\u00e1bio N. Demarqui (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A class of models for survival data with cure fraction and crossing survivals..<\/p>

Resumo:<\/strong>\u00a0In this talk, we introduce a new class of models to fit survival data with cure fraction and crossing survivals. The class of models proposed in this work has some attractive features: i) it is built upon a well-known unified two-stage process that possesses an appealing biological motivation in terms of incidence-latency of disease; ii) the incidence sub-model can be modeled by the Bernoulli, Poisson, negative binomial and Bell distributions; iii) the Yang and Prentice (YP) regression structure assumed to model the latency sub-model allows the model to accommodate survival data with crossing survivals, and it further includes the well-known proportional hazards (PH) and proportional odds (PO) models as particular cases; iv) the baseline survival distribution can be modeled parametrically (under the assumption of any parametric distribution), or semiparametrically (by either the piecewise exponential distribution or the Bernstein polynomials), providing greater flexibility for the modeling process; v) the likelihood function is available in closed-form expressions, leading to more straightforward inferential procedures. An extensive simulation study was carried out to investigate the asymptotic properties of the proposed class of models using the R package survcure, developed to fit the models belonging to the proposed class. We illustrate the usefulness of the proposed model through the analysis of a real dataset involving patients diagnosed with melanoma cancer previously investigated in the literature. The results obtained suggest that the proposed model arises as a flexible and attractive alternative to model survival data with cure fraction and crossing survivals.<\/p>

09\/09\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0sala 2076 – ICEx\/UFMG
<\/a><\/strong><\/p>

Fl\u00e1vio B. Gon\u00e7alves (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Beyond Gaussian processes: flexible Bayesian modeling and inference for geostatistical processes.<\/p>

Resumo:\u00a0<\/strong>In this talk, I will present a novel family of geostatistical models to account for features that cannot be properly accommodated by traditional Gaussian processes. The family is specified hierarchically, through a latent Poisson process, and combines the infinite-dimensional dynamics of Gaussian processes with that of any multivariate continuous distribution. The resulting process is called the Poisson-Gaussian Mixture Process – POGAMP. Whilst the attempt of defining geostatistical processes by assigning some arbitrary continuous distribution to be the finite-dimensional distributions usually leads to non-valid processes, the finite-dimensional distributions of the POGAMP can be arbitrarily close to any continuous distribution and still define a valid process. Formal results to establish the existence and some important properties of the POGAMP, such as absolute continuity with respect to a Gaussian process measure, are provided. Also, an MCMC algorithm is carefully devised to perform Bayesian inference when the POGAMP is discretely observed in some space domain.<\/p>

02\/09\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a>
<\/a><\/strong><\/p>

Simon Lunagomez (ITAM, M\u00e9xico)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Latent space modelling of hypergraph data.<\/p>

Resumo:\u00a0<\/strong>The increasing prevalence of relational data describing interactions among a target population has motivated a wide literature on statistical network analysis. In many applications, interactions may involve more than two members of the population and this data is more appropriately represented by a hypergraph. In this paper, we present a model for hypergraph data which extends the well-established latent space approach for graphs and, by drawing a connection to constructs from computational topology, we develop a model whose likelihood is inexpensive to compute. A delayed-acceptance MCMC scheme is proposed to obtain posterior samples and we rely on Bookstein coordinates to remove the identifiability issues associated with the latent representation. We theoretically examine the degree distribution of hypergraphs generated under our framework and, through simulation, we investigate the flexibility of our model and consider estimation of predictive distributions. Finally, we explore the application of our model to two real-world datasets. This is joint work with Kathryn Turnbull, Christopher Nemeth and Edoardo Airoldi.<\/p>

ANO DE 2022 – 1\u00ba SEMESTRE<\/strong><\/p>

15\/07\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0sala 2076 – ICEx\/UFMG
<\/a><\/strong><\/p>

Guilherme L. Oliveira (CEFET – MG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0An overview of Bayesian models for underreported count data: theory and applications.<\/p>

Resumo:\u00a0<\/strong>Count data is collected in many fields such as criminology, demography and epidemiology to assess or monitor the associated risks. In Brazil, this type of data usually comes from official registration systems which are prone to under-registration: only a fraction of the true (but unobserved) counts are reported. In this talk, some statistical approaches for correcting underreporting in count data will be discussed. The methods are based on the definition of a Poisson regression model for the observed data along with the specification of an auxiliary structure for modeling the reporting process. The inference is made under the Bayesian framework and it depends on the sort of prior information that is available. Applications consider Brazilian data on infant mortality, syphilis and tuberculosis, in which the correction of underreporting bias is very important for accurate surveillance, intervention and control by the government.<\/p>

08\/07\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0sala 2076 – ICEx\/UFMG
<\/a><\/strong><\/p>

Uriel M. Silva (Observat\u00f3rio de Sa\u00fade Urbana de BH, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A unified framework for sequential parameter learning with regularization in state space models.<\/p>

Resumo:\u00a0<\/strong>A unified framework for sequential parameter learning in state space models is proposed. This framework is capable of accommodating several other algorithms found in the literature as special cases, and this generality is achieved mainly by providing an alternative formalism to the role of regularization in this setting. In order to illustrate its flexibility, three novel algorithms are developed within this framework, including an improved and fully-adapted version of the celebrated Liu and West filter. These regularization techniques are associated with efficient resampling schemes, and their use is illustrated in challenging nonlinear settings with both synthetic and real-world data.<\/p>

01\/07\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Esther Salazar (FDA – Food and Drug Administration, EUA)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Flexible models for heterogeneous multiview data: applications to behavioral and fMRI data.<\/p>

Resumo:\u00a0<\/strong>We present a probabilistic framework for learning with heterogeneous multiview data where some views are given as ordinal, binary, or real-valued feature matrices, and some views as similarity matrices. Our framework has the following distinguishing aspects: (i) a unified latent factor model for integrating information from diverse feature (ordinal, binary, real) and similarity-based views, and predicting the missing data in each view, leveraging view correlations; (ii) seamless adaptation to binary\/multiclass classification where data consists of multiple feature and\/or similarity-based views; and (iii) an efficient, variational inference algorithm which is especially flexible in modeling the views with ordinal-valued data (by learning the cutpoints for the ordinal data), and extends naturally to streaming data settings. Our framework subsumes methods such as multiview learning and multiple kernel learning as special cases. We demonstrate the effectiveness of our framework on several real-world and benchmark datasets.<\/p>

24\/06\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0sala 2076 – ICEx\/UFMG<\/strong><\/p>

Douglas R. M. Azevedo (R\/Shiny developer, Appsilon)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Flexible link function with asymptotes: estimating the SUS population in Brazil.<\/p>

Resumo:\u00a0<\/strong>The estimation of hidden sub-populations is a hard task that appears in many fields. For example, public health planning in Brazil depends crucially on the number of people who holds a private health insurance plan and, hence, rarely uses the public services. Different sources of information about these sub-populations may be available at different geographical levels. The available information can be transferred between these different geographic levels to improve the estimation of the hidden population size. In this study, we propose a model that uses individual-level information to learn about the dependence between the response variable and explanatory variables by proposing a family of link functions with asymptotes that are flexible enough to represent the real aspects of the data and robust to departures from the model. We use the fitted model to estimate the size of the sub-population at any desired level. We illustrate our methodology by estimating the sub-population that uses the public health system in each neighborhood of large cities in Brazil.<\/p>

10\/06\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Artur J. Lemonte (Departamento de Estat\u00edstica, UFRN)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>On the local power of the LR, Wald, score e gradient tests under orthogonality.<\/p>

Resumo:\u00a0<\/strong>The local power of the LR, Wald, score e gradient tests under the presence of a parameter vector, omega say, that is orthogonal to the remaining parameters is studied. We show that some of the coefficients that define the local power of the tests remain unchanged regardless of whether omega is known or needs to be estimated, whereas the others can be written as the sum of two terms, the first of which being the corresponding term obtained as if omega were known, and the second, an additional term yielded by the fact that omega is unknown. We apply the general result in the class of nonlinear Student-t regression models.<\/p>

03\/06\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Frederico M. Almeida (P\u00f3s-doc, Escola de Nutri\u00e7\u00e3o, UFOP)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modified score function for monotone likelihood in the semiparametric mixture cure model.<\/p>

Resumo:\u00a0<\/strong>The cure fraction models are intended to analyze lifetime data from populations where some individuals are immune to the event under study, and allow a joint estimation of the distribution related to the cured and susceptible subjects, as opposed to the usual approach ignoring the cure rate. In situations involving small sample sizes with many censored times, the detection of non-finite coefficients may arise via maximum likelihood. This phenomenon is commonly known as monotone likelihood (ML), occurring in the Cox and logistic regression models when many categorical and unbalanced covariates are present. An existing solution to prevent the issue is based on the Firth correction, originally developed to reduce the estimation bias. The method ensures finite estimates by penalizing the likelihood function. In the context of mixture cure models, the ML issue is rarely discussed in the literature; therefore, this topic can be seen as the first contribution of our paper. The second major contribution, not well addressed elsewhere, is the study of the ML issue in cure mixture modeling under the flexibility of a semiparametric framework to handle the baseline hazard. We derive the modified score function based on the Firth approach and explore the finite sample size properties of the estimators via a Monte Carlo scheme. The simulation results indicate that the performance of coefficients related to the binary covariates are strongly affected by the imbalance degree. A real illustration (melanoma data) is discussed using a relatively novel data set collected in a Brazilian university hospital.<\/p>

27\/05\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Alessandro J. Q. Sarnaglia (Departamento de Estat\u00edstica, UFES)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Regress\u00e3o segmentada com abordagem Bayesiana para dados de contagem: Aplica\u00e7\u00e3o para estima\u00e7\u00e3o do limiar cr\u00edtico de polui\u00e7\u00e3o do ar em admiss\u00f5es hospitalares.<\/p>

Resumo:\u00a0<\/strong>A polui\u00e7\u00e3o do ar \u00e9 um problema enfrentado em v\u00e1rias partes do mundo. Em especial, como demonstrado por v\u00e1rios estudos, o material particulado com di\u00e2metro inferior a 10 \u00b5m (PM10) \u00e9 considerado um dos poluentes mais danosos \u00e0 sa\u00fade. Do ponto de vista de sa\u00fade p\u00fablica, frequentemente, esse impacto \u00e9 investigado por meio do estudo do efeito da concentra\u00e7\u00e3o do PM10 no n\u00famero de interna\u00e7\u00f5es hospitalares. Nesse sentido, o objetivo central deste trabalho \u00e9 realizar uma an\u00e1lise com foco em determinar a partir de qual n\u00edvel de concentra\u00e7\u00e3o de PM10 crian\u00e7as com 10 anos ou menos ficariam mais vulner\u00e1veis gerando, como consequ\u00eancia, um aumento no n\u00famero de admiss\u00f5es hospitalares por fatores respirat\u00f3rios. Para alcan\u00e7ar este objetivo, faremos uso de modelagem de regress\u00e3o segmentada sob o ponto de vista bayesiano. Como j\u00e1 pontuado na literatura, a verossimilhan\u00e7a nesse caso acaba n\u00e3o sendo diferenci\u00e1vel no ponto de quebra, o que se torna um desafio para m\u00e9todos que fazem uso de derivadas. Nesse sentido, propomos a utiliza\u00e7\u00e3o da aproxima\u00e7\u00e3o de Laplace para amostrar da distribui\u00e7\u00e3o a posteriori, recorrendo a uma reparametriza\u00e7\u00e3o do modelo e a m\u00e9todos bootstrap para especifica\u00e7\u00e3o da matriz de covari\u00e2ncias utilizada nessa aproxima\u00e7\u00e3o. Atrav\u00e9s de um estudo de simula\u00e7\u00e3o, comparamos esse m\u00e9todo a diferentes procedimentos j\u00e1 existentes na literatura, a fim de analisar a acur\u00e1cia das estimativas e o tempo computacional de execu\u00e7\u00e3o. Por meio dos resultados obtidos, conclu\u00edmos que a metodologia proposta apresenta resultados superiores \u00e0s metodologias existentes, j\u00e1 que a mesma obteve probabilidades de cobertura maiores se aproximando mais do valor de 95% de n\u00edvel de confian\u00e7a, al\u00e9m de apresentar maior precis\u00e3o com as amplitudes dos intervalos sendo menores. Por fim, aplicamos essa metodologia para estudar o efeito do PM10 e de vari\u00e1veis meteorol\u00f3gicas no n\u00famero de interna\u00e7\u00f5es di\u00e1rias por causas respirat\u00f3rias em um hospital do Esp\u00edrito Santo, Brasil. Como resultado, identificamos que o valor do limiar cr\u00edtico do poluente PM10 que acarretaria o aumento no n\u00famero de interna\u00e7\u00f5es infantis \u00e9 em torno de 34 \u00b5g\/m\u00b3, que est\u00e1 abaixo do referencial de 50 \u00b5g\/m\u00b3 estipulado pela Organiza\u00e7\u00e3o Mundial da Sa\u00fade (OMS). Resultado similar foi previamente obtido por Sarnaglia et al. (2021) sob o ponto de vista frequentista.<\/p>

20\/05\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local: sala 2076 – ICEx\/UFMG<\/strong><\/p>

Marcelo R. Hil\u00e1rio (Departamento de Matem\u00e1tica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Lei dos grandes n\u00fameros para passeios aleat\u00f3rios em ambientes aleat\u00f3rios din\u00e2micos<\/p>

Resumo:\u00a0<\/strong>Passeios aleat\u00f3rios em ambientes aleat\u00f3rios modelam o comportamento de uma part\u00edcula cujo movimento est\u00e1 sujeito \u00e0 influ\u00eancia de um meio desordenado. O n\u00facleo de transi\u00e7\u00e3o que governa o movimento do passeio aleat\u00f3rio depende de uma fam\u00edlia de vari\u00e1veis aleat\u00f3rias indexadas pelo espa\u00e7o chamada de ambiente aleat\u00f3rio. Esse ambiente pode ser est\u00e1tico, quando as vari\u00e1veis s\u00e3o mantidas constantes ou din\u00e2mico quando elas tamb\u00e9m evoluem estocasticamente no tempo. Nesta palestra vamos discutir alguns resultados recentes no entendimento do comportamento assint\u00f3tico do passeio no caso em que o ambiente \u00e9 din\u00e2mico. Em particular, ser\u00e1 apresentada uma t\u00e9cnica que permite demonstrar a lei dos grandes n\u00fameros para esses processos no caso em que o ambiente \u00e9 unidimensional e apresenta fortes correla\u00e7\u00f5es espa\u00e7o-temporais.<\/p>

13\/05\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Silvia L. P. Ferrari (Departamento de Estat\u00edstica, IME, USP)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Robust estimation in beta regression via maximum Lq-likelihood<\/p>

Resumo:\u00a0<\/strong>Beta regression models are widely used for modeling continuous data limited to the unit interval, such as proportions, fractions, and rates. The inference for the parameters of beta regression models is commonly based on maximum likelihood estimation. However, it is known to be sensitive to discrepant observations. In some cases, one atypical data point can lead to severe bias and erroneous conclusions about the features of interest. In this work, we develop a robust estimation procedure for beta regression models based on the maximization of a reparameterized Lq-likelihood. The new estimator offers a trade-off between robustness and efficiency through a tuning constant. To select the optimal value of the tuning constant, we propose a data-driven method that ensures full efficiency in the absence of outliers. We also improve on an alternative robust estimator by applying our data-driven method to select its optimum tuning constant. Monte Carlo simulations suggest marked robustness of the two robust estimators with little loss of efficiency when the proposed selection scheme for the tuning constant is employed. Applications to three datasets are presented and discussed. As a by-product of the proposed methodology, residual diagnostic plots based on robust fits highlight outliers that would be masked under maximum likelihood estimation. Joint work with Terezinha K. A. Ribeiro.<\/p>

06\/05\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Mark D. Risser (Lawrence Berkeley National Laboratory, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Bayesian inference for high-dimensional nonstationary Gaussian processes<\/p>

Resumo:\u00a0<\/strong>In spite of the diverse literature on nonstationary spatial modelling and approximate Gaussian process (GP) methods, there are no general approaches for conducting fully Bayesian inference for moderately sized nonstationary spatial data sets on a personal laptop. For statisticians and data scientists who wish to conduct posterior inference and prediction with appropriate uncertainty quantification, the lack of such approaches and software is a limitation. In this work, we develop methodology for implementing formal Bayesian inference for a general class of nonstationary GPs. Our novel approach uses pre-existing frameworks for characterizing nonstationarity in a new way while utilizing via modern GP likelihood approximations. Posterior sampling is implemented using flexible MCMC methods, with nonstationary posterior prediction conducted as a post-processing step. We demonstrate our novel methods on three data sets, ranging from several hundred to over several thousand locations. All of our methods are implemented in the freely available BayesNSGP software package for R.<\/p>

29\/04\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local: sala 2076 – ICEx\/UFMG<\/strong><\/p>

Marcelo A. Costa (Departamento de Engenharia de Produ\u00e7\u00e3o, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Dynamic time scan forecasting for multi-step wind speed prediction.<\/p>

Resumo:\u00a0<\/strong>Multi-step forecasting of wind speed time series, especially for day-ahead and longer time horizons, is still a challenging problem in the wind energy sector. In this paper, a novel analog-based methodology to perform multi-step forecasting in univariate time series, named dynamic time scan forecasting (DTSF), is presented. DTSF is a fast time series forecasting methodology for large data sets. Thus, the proposed method is optimal for forecasting renewable energy features such as wind speed, in which standard statistical and soft computing methods present limitations. A scan procedure is applied to identify similar patterns, named best matches, throughout the time series. As opposed to euclidean distance, more flexible similarity functions, using polynomial regression models, are dynamically estimated and Goodness-of-fit statistics are used to find the best matches. The observed values following the best matches and the fitted similarity functions are used to predict k-steps ahead, as well as forecasting intervals. An ensemble version of the method, named eDTSF, combines different predictions using different set of parameters thus, further improving forecasting performance. Remarkably, eDTSF achieved competitive results for multi-step forecasting of wind speed time series, even in situations of very high variability, as compared to eleven selected concurrent forecasting methods.<\/p>

08\/04\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local: sala 2076 – ICEx\/UFMG<\/strong><\/p>

Enrico A. Colosimo (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modelos de predi\u00e7\u00e3o cl\u00ednica.<\/p>

Resumo:\u00a0<\/strong>Modelos de predi\u00e7\u00e3o cl\u00ednica s\u00e3o constru\u00eddos com o objetivo de identificar pacientes ou indiv\u00edduos com maior probabilidade de desenvolver um espec\u00edfico evento, usualmente doen\u00e7a ou \u00f3bito. Estas predi\u00e7\u00f5es s\u00e3o utilizadas para mudar estilo de vida, guiar nas decis\u00f5es terap\u00eauticas, estratificar por gravidade, entre outros. Este trabalho foi motivado pela necessidade de construir um escore de risco para pacientes chag\u00e1sicos cardiopatas a partir de uma coorte acompanhada na regi\u00e3o do vale do Jequitinhonha, estado de Minas Gerais. Inicialmente foram obtidas predi\u00e7\u00f5es a partir linha de base, e a seguir, a medida que a coorte caminhou longitudinalmente, torn\u00e1-las din\u00e2micas. Vamos apresentar nesta palestra os passos fundamentais para a constru\u00e7\u00e3o de um escore de predi\u00e7\u00e3o est\u00e1tico e din\u00e2mico e ilustrar com os resultados obtidos para o estudo do vale do Jequitinhonha.<\/p>

25\/02\/2022 \u00e0s 14:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

C\u00e9sar Macieira (Doutorando, Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Clustering discrete data through the multinomial mixture model.<\/p>

Resumo:\u00a0<\/strong>Neste artigo, o modelo de mistura multinomial \u00e9 estudado atrav\u00e9s de uma abordagem de m\u00e1xima verossimilhan\u00e7a. \u00c9 apresentada a converg\u00eancia do estimador de m\u00e1xima verossimilhan\u00e7a para um conjunto com caracter\u00edsticas de interesse. M\u00e9todo este que visa selecionar o n\u00famero de componentes da mistura, desenvolvido com base na forma do estimador de m\u00e1xima verossimilhan\u00e7a. Em seguida, \u00e9 realizado um estudo de simula\u00e7\u00e3o para verificar seu comportamento. Por fim, duas aplica\u00e7\u00f5es em dados reais de misturas multinomiais s\u00e3o apresentadas. Refer\u00eancia: J. Portela (2008). Communications in Statistics – Theory and Methods, 37, 20, 3250-3263.<\/p>

25\/02\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Jonathan S. Matias (Doutorando, Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0M-Estimation in GARCH models.<\/p>

Resumo:\u00a0<\/strong>This paper derives asymptotic normality of a class of M-estimators in the generalized autoregressive conditional heteroskedastic (GARCH) model. The class of estimators includes least absolute deviation and Huber’s estimator in addition to the well-known quasi maximum likelihood estimator. For some estimators, the asymptotic normality results are obtained only under the existence of fractional unconditional moment assumption on the error distribution and some mild smoothness and moment assumptions on the score function. Reference: K. Mukherjee (2008). Econometric Theory, 24, 6, 1530-1553.<\/p>

18\/02\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Ricardo Cunha Pedroso (Doutorando, Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Dependent modeling of temporal sequences of random partitions.<\/p>

Resumo:\u00a0<\/strong>O semin\u00e1rio consistir\u00e1 na apresenta\u00e7\u00e3o do artigo “Dependent Modeling of Temporal Sequences of Random Partitions”, Page et al. (2021), onde os autores prop\u00f5em uma modelagem para sequ\u00eancias de parti\u00e7\u00f5es aleat\u00f3rias dependentes, no caso em que o principal interesse \u00e9 a identifica\u00e7\u00e3o de clusters. S\u00e3o apresentadas as propriedades condicionais e marginais do modelo conjunto das parti\u00e7\u00f5es e estrat\u00e9gias computacionais Bayesianas de estima\u00e7\u00e3o. Um estudo com dados simulados para o caso de depend\u00eancia temporal demonstra que o modelo produz estimativas para as parti\u00e7\u00f5es que evoluem de forma suave e, por fim, o modelo \u00e9 aplicado a dados de meio ambiente que exibem depend\u00eancia espa\u00e7o-temporal. Refer\u00eancia: Page G.L., Quintana F.A., Dahl D.B. (2021). Journal of Computational and Graphical Statistics, doi 10.1080\/10618600.2021.1987255.<\/p>

11\/02\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Gabriel Oliveira Assun\u00e7\u00e3o (Doutorando, Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Data augmentation approaches for NLP.<\/p>

Resumo:\u00a0<\/strong>Recentemente o interesse em Data Augmentation na \u00e1rea Processamento de Linguagem Natural (NLP) aumentou devido a trabalhos em dom\u00ednios de pouco recurso, novos tipos de tarefas e a popularidade em redes neurais de larga escala que necessitam de uma quantidade grande de dados para ser treinada. Mesmo ocorrendo este interesse na \u00e1rea, ela ainda \u00e9 pouco explorada. Nesta apresenta\u00e7\u00e3o ser\u00e3o apresentados alguns m\u00e9todos existentes na literatura para Data Augmentation em NLP, com suas aplica\u00e7\u00f5es e desafios. Refer\u00eancia: Feng S.Y., Gangal V., Wei J., Chandar S., Vosoughi S., Mitamura T., Hovy E. (2021). A survey of data augmentation approaches for NLP. In Findings of the Association for Computational Linguistics: ACL-IJCNLP 2021, 968-988, Association for Computational Linguistics.<\/p>

04\/02\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Gisele de Oliveira Maia (Doutoranda, Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Observation-driven models for Poisson counts.<\/p>

Resumo:<\/strong>\u00a0O semin\u00e1rio \u00e9 focado em apresentar o artigo Observation-driven models for Poisson counts, Davis et al. (2003), onde \u00e9 abordado o Modelo Linear Generalizado Autorregressivo M\u00e9dias-M\u00f3veis para s\u00e9ries temporais de contagens, cuja distribui\u00e7\u00e3o condicional da s\u00e9rie temporal dada suas observa\u00e7\u00f5es passadas e covari\u00e1veis segue a distribui\u00e7\u00e3o Poisson. Ser\u00e3o apresentadas a estrutura, propriedades, m\u00e9todo de estima\u00e7\u00e3o dos par\u00e2metros e propriedades assint\u00f3ticas deste modelo. Simula\u00e7\u00f5es s\u00e3o apresentadas com o intuito de fornecer informa\u00e7\u00f5es adicionais sobre o comportamento dos estimadores. Finalmente, \u00e9 descrita uma aplica\u00e7\u00e3o a um modelo de regress\u00e3o para contagens di\u00e1rias de casos de asma em um hospital de Sydney, Austr\u00e1lia. Refer\u00eancia: Davis R.A., Dunsmuir W.T.M., Streett S.B. (2003) Observation\u2010driven models for Poisson counts. Biometrika, 90, 4, 777-790.<\/p>

28\/01\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Camila B. Zeller (Departamento de Estat\u00edstica, UFJF)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Estimation in the multivariate linear regression models with skew scale mixtures of normal distributions..<\/p>

Resumo:<\/strong>\u00a0In this paper, we present recent results in the context of multivariate linear regression models considering that random errors follow multivariate skew scale mixtures of normal distributions. This class of distributions includes the scale mixtures of multivariate normal distributions, as special cases, and provides flexibility in capturing a wide variety of asymmetric behaviors. We implemented the algorithm ECM (Expectation\/Conditional Maximization) and we obtained closed-form expressions for all the estimators of the parameters of the proposed model. The proposed algorithm and methods are implemented in the new R package skewMLRM. Finally, a real data set is analyzed in order to show the usefulness of the package.<\/p>

21\/01\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Silvana Schneider (IME, UFRGS)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0An approach for long-term survival data with dependent censoring.<\/p>

Resumo:<\/strong>\u00a0In this paper, we propose a likelihood-based approach for long-term multivariate survival data, which is suitable to accommodate the dependent censoring. The association between lifetimes and dependent censoring is accommodated through the conditional approach of the frailty models. The marginal distributions can be adjusted assuming Weibull or piecewise exponential (PE) distributions. A Monte Carlo Expectation-Maximization algorithm is developed to estimate the proposed estimators. The simulation study results show a small relative bias and coverage probability near the nominal value. Finally, in order to evaluate the life dynamic of free-ranging dogs, taking into account all characteristics of the data, including long-term survival, we analyze the survival times of stray dogs in India).<\/p>

14\/01\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Rosangela H. Loschi (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Handling categorical features with many levels using a product partition model.<\/p>

Resumo:<\/strong>\u00a0A common difficulty in data analysis is how to handle categorical predictors with a large number of levels or categories. Few proposals have been developed to tackle this important and frequent problem. We introduce a generative model that simultaneously carries out the model fitting and the aggregation of the categorical levels into larger groups. We represent the categorical predictor by a graph where the nodes are the categories and establish a probability distribution over meaningful partitions of this graph. Conditionally on the observed data, we obtain a posterior distribution for the levels aggregation, allowing the inference about the most probable clustering for the categories. Simultaneously, we extract inference about all the other regression model parameters. We compare our and state-of-art methods showing that it has equally good predictive performance and more interpretable results. Our approach balances out accuracy versus interpretability, a current important concern in statistics and machine learning. Joint work with: Tulio Criscuolo (Google-USA), Renato Assun\u00e7\u00e3o (ESRI, USA), Wagner Meira (DCC, UFMG) and Danna Cruz (Universidad del Rosario, Co).<\/p>

07\/01\/2022 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

P. Richard Hahn (Arizona State University, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Feature selection for causal effect estimation.<\/p>

Resumo:\u00a0<\/strong>This paper defines the notion of a minimal control function, on the basis of which a novel regression penalty is devised that is unbiased for average treatment effects. The development of the new approach combines insights from three distinct methodological traditions for studying causal effect estimation: potential outcomes, causal diagrams, and structural models with additive errors. It is demonstrated that naive feature selection and\/or regularization approaches to treatment effect estimation can exhibit severe bias for average and conditional average treatment effects.<\/p>

.ANO DE 2021 – 2\u00ba SEMESTRE<\/strong><\/p>

17\/12\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Roger W. C. Silva (Departamento de Estat\u00edstica – UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Constrained-degree percolation in random environment.<\/p>

Resumo:\u00a0<\/strong>We consider the Constrained-degree percolation model in random environment on the square lattice. In this model, each vertex v has an independent random constraint \u03ba_v which takes the value j \u2208 {0, 1, 2, 3} with probability \u03c1_j . Each edge e attempts to open at a random uniform time U_e in [0, 1], independently of all other edges. It succeeds if at time U_e both its end-vertices have degrees strictly smaller than their respectively attached constraints. We show that this model undergoes a non-trivial phase transition when \u03c1_3 is sufficiently large. The proof consists of a decoupling inequality, the continuity of the probability for local events, and a coarse-graining argument. Joint work with Diogo Santos and R\u00e9my Sanchis.<\/p>

10\/12\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Airlane P. Alencar (Departamento de Estat\u00edstica, IME, USP).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modelos GARMA para s\u00e9ries temporais \u2013 GARMA modificado e outras distribui\u00e7\u00f5es.<\/p>

Resumo:\u00a0<\/strong>Em muitos problemas reais, queremos analisar se h\u00e1 mudan\u00e7as de tend\u00eancia, sazonalidade e efeitos de covari\u00e1veis em s\u00e9ries temporais. Podemos considerar modelos de regress\u00e3o linear e modelos lineares generalizados, levando em conta a autocorrela\u00e7\u00e3o, ajustando os modelos de regress\u00e3o com erros SARMA e modelos GARMA (Benjamin et al. 2003). Devido \u00e0 multicolineariedade, propomos um modelo GARMA modificado (Albarracin et al. 2019). Considerando outras distribui\u00e7\u00f5es, os modelos GARMA usuais, como a Conway-Maxwell Poisson (Melo e Alencar, 2020), que admitem super, sub e equidispers\u00e3o. Trabalho em conjunto com: Orlando Y.E. Albarracin (IME-USP), Moizes Melo (UFRN) e Linda Lee Ho (EP-USP).<\/p>

03\/12\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

RHelton Saulo B. Santos (Departamento de Estat\u00edstica, UnB).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modelos autorregressivos de dura\u00e7\u00e3o condicional.<\/p>

Resumo:\u00a0<\/strong>Modelos autorregressivos de dura\u00e7\u00e3o condicional (ACD) s\u00e3o utilizados principalmente para lidar com dados de dura\u00e7\u00e3o de transa\u00e7\u00f5es financeiras. Tais dados possuem informa\u00e7\u00f5es \u00fateis sobre as atividades do mercado. Nesta apresenta\u00e7\u00e3o, o modelo original ACD e algumas variantes s\u00e3o apresentadas..<\/p>

26\/11\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Rodrigo Lambert (FAMAT, UFU).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A fun\u00e7\u00e3o de sobreposi\u00e7\u00e3o no contexto da recorr\u00eancia de Poincar\u00e9.<\/p>

Resumo:\u00a0<\/strong>O estudo de tempos de primeiro retorno e fun\u00e7\u00e3o de sobreposi\u00e7\u00e3o em sequ\u00eancias simb\u00f3licas tem forte apelo em aplica\u00e7\u00f5es. Desde gen\u00e9tica com o estudo de sequ\u00eancias de DNA at\u00e9 a teoria da informa\u00e7\u00e3o com o estudo de algoritmos de compress\u00e3o, tal assunto se mostra uma ferramenta potencialmente \u00fatil para atacar problemas que pertencem a diferentes \u00e1reas do conhecimento. Nessa apresenta\u00e7\u00e3o, come\u00e7arei motivando e dando as defini\u00e7\u00f5es de tempo de primeiro retorno e fun\u00e7\u00e3o de sobreposi\u00e7\u00e3o, e comentarei alguns resultados conhecidos da literatura. Finalmente, apresentarei um resultado recentemente obtido com E. A. Rada-Mora (UFABC).<\/p>

19\/11\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Vera L\u00facia Damasceno Tomazella (Departamento de Estat\u00edstica, UFSCar).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Nonproportional hazards model with a frailty term for modeling subgroups with evidence of long-term survivors: Application to a lung cancer dataset..<\/p>

Resumo:\u00a0<\/strong>With advancements in medical treatments for cancer, an increase in the life expectancy of patients undergoing new treatments is expected. Consequently, the field of statistics has evolved to present increasingly flexible models to explain such results better. In this paper, we present a lung cancer dataset with some covariates that exhibit nonproportional hazards (NPHs). Besides, the presence of long\u2010term survivors is observed in subgroups. The proposed modeling is based on the generalized time\u2010dependent logistic model with each subgroup’s effect time and a random term effect (frailty). In practice, essential covariates are not observed for several reasons. In this context, frailty models are useful in modeling to quantify the amount of unobservable heterogeneity. The frailty distribution adopted was the weighted Lindley distribution, which has several interesting properties, such as the Laplace transform function on closed form, flexibility in the probability density function, among others. The proposed model allows for NPHs and long\u2010term survivors in subgroups. We exemplify this model’s use by applying data of patients diagnosed with lung cancer in the state of S\u00e3o Paulo, Brazil.<\/p>

12\/11\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Jo\u00e3o Batista de Morais Pereira (DME, UFRJ).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Spatial confounding in hurdle multilevel beta models: the case of the Brazilian Mathematical Olympics for Public Schools.<\/p>

Resumo:\u00a0<\/strong>Among the many disparities for which Brazil is known is the difference in performance across students who attend the three administrative levels of Brazilian public schools: federal, state and municipal. Our main goal is to investigate whether student performance in the Brazilian Mathematical Olympics for Public Schools is associated with school administrative level and student gender. For this, we propose a hurdle hierarchical beta model for the scores of students who took the examination in the second phase of these Olympics, in 2013. The mean of the beta model incorporates fixed and random effects at the student and school levels. We explore different distributions for the random school effect. As the posterior distributions of some fixed effects change in the presence, and distribution, of the random school effects, we also explore models that constrain random school effects to the orthogonal complement of the fixed effects. We conclude that male students perform slightly better than female students and that, on average, federal schools perform substantially better than state or municipal schools. However, some of the best municipal and state schools perform as well as some federal schools. We hypothesize that this is due to individual teachers who successfully motivate and prepare their students to perform well in the mathematical Olympics. Joint work with Widemberg Nobre, Igor Silva and Alexandra Schmidt..<\/p>

05\/11\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Fernanda De Bastiani (Departamento de Estat\u00edstica, UFPE).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Regress\u00e3o flex\u00edvel com GAMLSS.<\/p>

Resumo:\u00a0<\/strong>Os GAMLSS (Generalized Additive Models for Location, Scale, and Shape) podem ser considerados como ferramenta de regress\u00e3o apropriada para conjuntos de dados onde a distribui\u00e7\u00e3o da vari\u00e1vel resposta pode ser uma distribui\u00e7\u00e3o param\u00e9trica muito flex\u00edvel (al\u00e9m de pertencente \u00e0 fam\u00edlia exponencial) e onde todos os par\u00e2metros da distribui\u00e7\u00e3o (n\u00e3o apenas a m\u00e9dia) podem ser modelados usando ou fun\u00e7\u00f5es suaves das vari\u00e1veis explicativas. GAMLSS fornece uma estrutura para abordar problemas como a escolha de uma distribui\u00e7\u00e3o apropriada para a vari\u00e1vel resposta e explicando como essa distribui\u00e7\u00e3o, e seus par\u00e2metros, variam em diferentes valores das vari\u00e1veis explicativas. considera diferentes termos aditivos para modelar os par\u00e2metros da distribui\u00e7\u00e3o, como linear, suaviza\u00e7\u00e3o n\u00e3o param\u00e9trica e termos de efeitos aleat\u00f3rios. E cont\u00e9m diferentes t\u00e9cnicas de sele\u00e7\u00e3o de modelagem e diagn\u00f3sticos para verificar a adequa\u00e7\u00e3o do modelo tamb\u00e9m ser\u00e3o abordados.<\/p>

29\/10\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Caio Lucidius Naberezny Azevedo (Departamento de Estat\u00edstica, Unicamp).<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Bayesian longitudinal item response modeling with multivariate asymmetric serial dependencies.<\/p>

Resumo:\u00a0<\/strong>It is usually impossible to impose experimental conditions in large-scale longitudinal (observational) studies in education. This increases the risk of bias due to for instance unobserved heterogeneity, missing background variables, and dropouts. A flexible statistical model is required for the nature of the observational assessment data and to account for the unexplained heterogeneity. A general class of longitudinal item response theory (IRT) models is proposed, where growth in performance can be monitored using a skewed multivariate normal distribution for the latent variables. Change in performance and unexplained heterogeneity is addressed through structured covariance patterns and skewed multivariate latent variable distributions. The Cholesky decomposition of the covariance matrix is considered to model the dependence structure. A novel multivariate skewnormal distribution is defined by the antedependence model with centered skew-normal distributed errors. A hybrid MCMC approach is developed for parameter estimation, model-fit assessment, and model comparison. Results of simulation studies show good parameter recovery. A longitudinal assessment study by the Brazilian federal government is considered to show the performance of the general LIRT model. Joint work with: Jos\u00e9 Roberto S. Santos (Department of Statistics and Applied Mathematics, Federal University of Cear\u00e1 – Brazil) and Jean-Paul Fox (Department of Research Methodology, Measurement and Data Analysis, University of Twente – Netherlands).<\/p>

22\/10\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Wagner Hugo Bonat (Departamento de Estat\u00edstica, UFPR))<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Multivariate covariance generalized linear models with applications in R..<\/p>

Resumo:\u00a0<\/strong>In this talk I will present a recent proposed framework for non-normal multivariate data analysis called multivariate covariance generalized linear models (McGLMs), designed to handle multivariate response variables, along with a wide range of temporal and spatial correlation structures defined in terms of a generalized Kronecker product. The models take non-normality into account in the conventional way by means of a variance function, and the mean structure is modelled by means of a link function and a linear predictor. The covariance structure is modelled by means of a covariance link function combined with a matrix linear predictor involving known matrices. The models are fitted using an efficient Newton scoring algorithm based on quasi-likelihood and Pearson estimating functions, using only second-moment assumptions. McGLMs provide a unified approach to a wide variety of different types of response variables and covariance structures, including multivariate extensions of repeated measures, time series, longitudinal, spatial and spatio-temporal data. Furthermore, I present the computational implementation in R through the package mcglm. Illustrations include mixed models, longitudinal data analysis, spatial models for areal data, models to deal with mixed outcomes and multivariate models for count data using the Poisson-Tweedie distribution.<\/p>

ANO DE 2021 – 1\u00ba SEMESTRE<\/strong><\/p>

10\/09\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Alisson Carlos da Costa Silva (Doutorando em Estat\u00edstica, DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Cure frailty models for survival data: Application to recurrences for breast cancer and to hospital readmissions for colorectal cancer.<\/p>

Resumo:\u00a0<\/strong>Owing to the natural evolution of a disease, several events often arise after a first treatment for the same subject. For example, patients with a primary invasive breast cancer and treated with breast conserving surgery may experience breast cancer recurrences, metastases or death. A certain proportion of subjects in the population who are not expected to experience the events of interest are considered to be \u2018cured\u2019 or non-susceptible. To model correlated failure time data incorporating a surviving fraction, we compare several forms of cure rate frailty models. In the first model already proposed non-susceptible patients are those who are not expected to experience the event of interest over a sufficiently long period of time. The other proposed models account for the possibility of cure after each event. We illustrate the cure frailty models with two data sets. First to analyse time-dependent prognostic factors associated with breast cancer recurrences, metastases, new primary malignancy and death. Second to analyse successive rehospitalizations of patients diagnosed with colorectal cancer. Estimates were obtained by maximization of likelihood using SAS proc NLMIXED for a piecewise constant hazards model. As opposed to the simple frailty model, the proposed methods demonstrate great potential in modelling multivariate survival data with long-term survivors (‘cured’ individuals). Refer\u00eancia: Rondeau V, Schaffner E, Corbiere F, Gonzalez JR & Mathoulin-P\u00e9lissier S (2013), Statistical Methods in Medical Research, 22, 3, 243-260.<\/p>

03\/09\/2021,\u00a0excepcionalmente \u00e0s 15:00 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Ming-Hui Chen (University of Connecticut, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A power prior approach for leveraging external longitudinal and competing risks survival data within the joint modeling framework.<\/p>

Resumo:\u00a0<\/strong>In this paper, we propose a new partial borrowing-by-parts power prior for carrying out the analysis of co-longitudinal and survival data within the joint modeling framework. The borrowing-by-parts power prior facilitates borrowing the information from a subset of the data, from a subset of the model parameters, or from the different parts of the joint model. The deviance information criterion is used to quantify the gain in the fit of the current longitudinal and survival data when leveraging external co-data. A Markov chain Monte Carlo sampling algorithm is developed for carrying out Bayesian computations. The proposed methodology is motivated by two large concurrent clinical trials: Selenium and Vitamin E Cancer Prevention Trial (SELECT) and Prostate, Lung, Colon, Ovarian (PLCO) prevention trial. In both trials, the longitudinal biomarkers and competing risks survival data were collected. A detailed analysis of the PLCO and SELECT data is conducted to demonstrate the usefulness of the proposed methodology. This is a joint work with Md. Tuhin Sheikh, Jonathan A. Gelfond, and Joseph G. Ibrahim.<\/p>

27\/08\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Jun Yan (University of Connecticut, EUA)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Brownian motion governed by telegraph process in modeling high-frequency financial series..<\/p>

Resumo:\u00a0<\/strong>The classic Markov regime-switching model is a discrete-time model, which cannot naturally handle irregularly spaced time series. We propose a continuous-time regime-switching model with two states. In each state, the process is a Brownian motion with state-specific drift and volatility. The unobserved states are characterized by a telegraph process with exponential holding times, which is a continuous-time Markov process. Inferences for the model parameters with discretely spaced time series are developed on the basis of the hidden Markov model. Closes-form expressions for the likelihood are facilitated with the dynamic programming technique along with occupation time results for telegraph processes. For high-frequency data, a fast approximation reduces the computing time drastically without much accuracy loss. The performance of the method is validated in a simulation study. In application to a collection of stock prices, the model is found to be competitive in comparison to the popular GARCH model.<\/p>

20\/08\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Thais C. O. Fonseca (DME – IM, UFRJ)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Can you render your Lattes? A Bayesian Network modelling of digital preservation risks.<\/p>

Resumo:\u00a0<\/strong>Digital records comprise primary sources which may be physical, born-digital or digitised. They are under threat from rapidly evolving technology, outdated policies, and a skills gap across the archives sector. Thus, the preservation of digital material is a challenge for which many archives feel underprepared and ill-equipped. This talk presents the results of the Safeguarding the Nation\u2019s Memory Project which aimed to help archivists manage digital preservation risks through the creation of a new quantitative risk management framework. This project has produced the web-based app DiAGRAM (the Digital Archiving Graphical Risk Assessment Model) which quantifies the effect on preservation risk of various actions and interventions. This work brings Bayesian Network methods into the digital heritage sphere for the first time through close collaboration with specialists in this field. Soft elicitation was used to identify the most likely elements contributing to digital preservation and their interrelations. Where good quality data was not available, expert elicitation based on the IDEA protocol was applied.<\/p>

13\/08\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Fernando F. Nascimento (Depto. de Estat\u00edstica, UFPI)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modelo de regress\u00e3o para cauda e n\u00e3o-cauda de modelos de excessos, aplicado em dados de temperaturas m\u00e1ximas e m\u00ednimas.<\/p>

Resumo:\u00a0<\/strong>A rela\u00e7\u00e3o de ocorr\u00eancias ligadas \u00e0s altera\u00e7\u00f5es clim\u00e1ticas significativas t\u00eam crescido nos \u00faltimos anos. Essas altera\u00e7\u00f5es podem ser influenciadas por um conjunto de covari\u00e1veis, como temperatura, localiza\u00e7\u00e3o e tempo em que ocorrem. Analisar a rela\u00e7\u00e3o existente entre elementos e fatores que influenciam no comportamento de tais eventos \u00e9 de extrema relev\u00e2ncia para a tomada de decis\u00f5es com a finalidade de minimizar e at\u00e9 mesmo evitar poss\u00edveis danos e perdas. Este trabalho \u00e9 uma extens\u00e3o do modelo proposto por Behrens et al. (2004) que considera uma distribui\u00e7\u00e3o GPD para a cauda e uma distribui\u00e7\u00e3o Gama para n\u00e3o cauda, do modelo de Nascimento (2012) que combina a Distribui\u00e7\u00e3o de Pareto Generalizada (GPD) para dados acima de um limiar e mistura de Gamas para valores abaixo do limiar, e o modelo de Nascimento et al. (2011) que utiliza estrutura de regress\u00e3o para an\u00e1lise de valores extremos em todos os par\u00e2metros da cauda. A partir dos dados de temperaturas m\u00e1ximas em cidades dos Estados Unidos e temperaturas m\u00ednimas em cidades do Estado do Rio de Janeiro este trabalho foi conduzido com o objetivo de incorporar uma estrutura de regress\u00e3o para os par\u00e2metros de toda a distribui\u00e7\u00e3o, incluindo tamb\u00e9m os par\u00e2metros da distribui\u00e7\u00e3o abaixo da cauda. O modelo proposto consiste em uma distribui\u00e7\u00e3o Gama para a estima\u00e7\u00e3o dos valores abaixo do limiar e distribui\u00e7\u00e3o GPD para valores acima do limiar. A estima\u00e7\u00e3o dos par\u00e2metros ocorreu por meio de t\u00e9cnicas MCMC – Markov Chain Monte Carlo. Este modelo apresenta a vantagem de capturar comportamentos caracter\u00edsticos de todas as localiza\u00e7\u00f5es e \u00e9pocas do ano e fornecer melhor poder preditivo das estima\u00e7\u00f5es de medidas importantes em valores extremos como a estima\u00e7\u00e3o de quantis extremos.<\/p>

06\/08\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Michelle F. Miranda (University of Victoria, Canad\u00e1)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A computationally scalable Bayesian method for simultaneous detection of activation signatures and background connectivity for task fMRI data.<\/p>

Resumo:\u00a0<\/strong>Task-based functional magnetic resonance imaging (fMRI) studies are a powerful tool to understand human sensory, cognitive, and emotional processes. To optimally perform a task, the brain enters a task state, and it needs to maintain it throughout the task. It is hypothesized that this is done by brain modulation of task-dependent connection patterns. We will use the term “background connectivity” for the task-dependent modulations that are due to variations in ongoing brain activity instead of stimulus-driven activity. We propose a unified modelling approach to estimate activation signatures and background connectivity in the working-memory task of the Human Connectome Project. Our model involves a new hybrid tensor spatial-temporal basis strategy that enables scalable computing, yet it captures nearby and distant intervoxel correlation and long-memory temporal correlation. The spatial basis is a composite hybrid transform with two levels: the first accounts for within-ROI correlation, and the second between-ROI distant correlation. Our basis space model increases sensitivity for identifying activation signatures, partly driven by the induced background connectivity that itself can be summarized to reveal biological insights.<\/p>

30\/07\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Luis Mauricio Castro Cepero (PUC, Chile)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Modelling point referenced spatial count data: a Poisson process approach.<\/p>

Resumo:\u00a0<\/strong>Random fields are useful mathematical tools for representing natural phenomena with complex dependence structures in space and\/or time. In particular, the Gaussian random field is commonly used due to its attractive properties and mathematical tractability. However, this assumption seems to be restrictive when dealing with counting data. To deal with this situation, we propose a random field with a Poisson marginal distribution by considering a sequence of independent copies of a random field with an exponential marginal distribution as ‘inter-arrival times’ in the counting renewal processes framework. Our proposal can be viewed as a spatial generalization of the Poisson process. Unlike the classical hierarchical Poisson Log-Gaussian model, our proposal generates a (non)-stationary random field that is mean square continuous and with Poisson marginal distributions. For the proposed Poisson spatial random field, analytic expressions for the covariance function and the bivariate distribution are provided. In an extensive simulation study, we investigate the weighted pairwise likelihood as a method for estimating the Poisson random field parameters. Finally, the effectiveness of our methodology is illustrated by an analysis of reindeer pellet-group survey data, where a zero-inflated version of the proposed model is compared with zero-inflated Poisson Log-Gaussian and Poisson Gaussian copula models.<\/p>

23\/07\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Leonardo Soares Bastos (Fiocruz, Rio de Janeiro)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Nowcasting COVID-19 deaths and hospitalized cases in Brazil<\/p>

Resumo:<\/strong>\u00a0The coronavirus disease (COVID-19) pandemic continues to cause a massive burden in the world, especially in countries such as Brazil, with poor implementation of strategies to mitigate the transmission of SARS-CoV-2. The number of cases, severe cases, and deaths by COVID-19 are important indicators of how the COVID-19 epidemic is affecting a particular region and can be used by decision-makers to act in order to reduce morbidity and mortality. However, a common problem with surveillance data is reporting delays, whereby cases and deaths are recorded in the surveillance system days or even weeks after they occurred. Statistical models can estimate the actual number of cases, severe cases, and deaths by COVID-19 accounting for the delays (nowcasting). We proposed a Bayesian hierarchical model to nowcast deaths and hospitalised cases for Brazil and also for the 27 federal units. Finally, we provide some general discussion about the COVID-19 situation in Brazil.<\/p>

16\/07\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Marina Silva Paez (DME – UFRJ, Rio de Janeiro)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Anisotropia atrav\u00e9s de deforma\u00e7\u00e3o espacial em diferentes modelos geoestat\u00edsticos espa\u00e7o-temporais.<\/p>

Resumo:<\/strong>\u00a0Neste semin\u00e1rio irei apresentar diferentes classes de modelos geostat\u00edsticos que lidam com anisotropia por meio de processos de deforma\u00e7\u00e3o. Em suma, a ideia do procedimento de deforma\u00e7\u00e3o espacial consiste em fazer uma transforma\u00e7\u00e3o de R\u00b2 em R\u00b2 que mapeia as coordenadas geogr\u00e1ficas da regi\u00e3o de interesse S (possivelmente anisotr\u00f3pica) para um novo espa\u00e7o latente D (isotr\u00f3pico por constru\u00e7\u00e3o). A 1\u00aa proposta \u00e9 a de um modelo geoestat\u0131\u0301stico para fen\u00f4menos espa\u00e7o-temporais univariados que n\u00e3o s\u00e3o estacion\u00e1rios e exibem observa\u00e7\u00f5es at\u00edpicas. Propomos a modelagem atrav\u00e9s de um processo t-Student para descrever dados com caudas pesadas, com componentes espaciais e temporais separ\u00e1veis. A varia\u00e7\u00e3o no tempo \u00e9 incorporada atrav\u00e9s de modelos din\u00e2micos e a componente puramente espacial assume depend\u00eancia atrav\u00e9s da especifica\u00e7\u00e3o de uma fun\u00e7\u00e3o de correla\u00e7\u00e3o espacial. Lidamos com a anisotropia atrav\u00e9s de deforma\u00e7\u00e3o espacial de Sampson e Guttorp (1992), e, uma vez que adotamos o paradigma Bayesiano, nos baseamos na abordagem de Schmidt e O’Hagan (2003). A 2\u00aa proposta trata de modelos espa\u00e7o-temporais multivariados. Nos baseamos na modelagem proposta por Paez et al. (2008) que apresenta uma classe de modelos din\u00e2micos hier\u00e1rquicos para observa\u00e7\u00f5es matriz-variadas (no caso a matriz considera as dimens\u00f5es espa\u00e7o e tempo). Modelos din\u00e2micos s\u00e3o mais uma vez propostos para tratar de varia\u00e7\u00f5es temporais. Com o objetivo de relaxar a hip\u00f3tese de isotropia assumida no referido trabalho, a presente pesquisa prop\u00f5e uma extens\u00e3o para o trabalho de Paez et al. (2008) que permite acomodar superf\u00edcies anisotr\u00f3picas. A infer\u00eancia, como j\u00e1 mencionado, \u00e9 feita sob o ponto de vista Bayesiano, e propomos o uso do MCMC para amostrar da distribui\u00e7\u00e3o a posteriori dos par\u00e2metros dos modelos. As modelagens s\u00e3o inicialmente testadas para dados simulados e posteriormente aplicadas a conjuntos de dados ambientais. Colaboradores: Fidel E. C. Morales, Dimitris Politis, Jacek Leskow e Rodrigo Bulh\u00f5es.<\/p>

09\/07\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Daniel Takata Gomes (ENCE – IBGE, Rio de Janeiro)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Usain Bolt x Michael Phelps: c\u00e1lculo de \u00edndice de desempenho em esportes baseado em teoria de valores extremos.<\/p>

Resumo:<\/strong>\u00a0A Federa\u00e7\u00e3o Internacional de Nata\u00e7\u00e3o (FINA) utiliza um sistema de pontos que permite compara\u00e7\u00f5es de resultados de diferentes provas. Tal sistema \u00e9 importante por v\u00e1rias raz\u00f5es, pois \u00e9 utilizado como crit\u00e9rio para atribui\u00e7\u00e3o de pr\u00eamios em competi\u00e7\u00f5es e para forma\u00e7\u00e3o de sele\u00e7\u00f5es nacionais. Os pontos s\u00e3o atribu\u00eddos tendo como refer\u00eancia somente os recordes mundiais das provas oficiais. Neste trabalho \u00e9 sugerido um novo \u00edndice, baseado na distribui\u00e7\u00e3o de probabilidade das marcas dos nadadores mais r\u00e1pidos da hist\u00f3ria de cada prova. Pela Teoria de Valores Extremos, tal distribui\u00e7\u00e3o, sob certas condi\u00e7\u00f5es, converge para uma distribui\u00e7\u00e3o de Pareto generalizada. As compara\u00e7\u00f5es s\u00e3o feitas baseadas nas probabilidades de exced\u00eancia relativas \u00e0s marcas dos nadadores. Tamb\u00e9m \u00e9 feita uma compara\u00e7\u00e3o de desempenhos de esportistas de diferentes modalidades, no caso atletismo e nata\u00e7\u00e3o, com o objetivo de avaliar quem obteve o resultado mais extremo entre dois dos maiores nomes da hist\u00f3ria do esporte: o jamaicano Usain Bolt e o americano Michael Phelps.<\/p>

02\/07\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Florencia Leonardi (IME – USP, S\u00e3o Paulo)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Detec\u00e7\u00e3o de estrutura de intera\u00e7\u00e3o para campos Markovianos discretos sobre grafos..<\/p>

Resumo:<\/strong>\u00a0Os campos aleat\u00f3rios de Markov discretos sobre grafos, tamb\u00e9m conhecidos como modelos gr\u00e1ficos na literatura estat\u00edstica, t\u00eam se popularizado nos \u00faltimos anos devido \u00e0 sua flexibilidade para capturar rela\u00e7\u00f5es de depend\u00eancia condicional entre vari\u00e1veis. Eles j\u00e1 foram aplicados a muitos problemas diferentes em campos diferentes, como Biologia, Ci\u00eancias Sociais ou Neuroci\u00eancias. Os modelos gr\u00e1ficos s\u00e3o, em certo sentido, vers\u00f5es “finitas” de campos aleat\u00f3rios gerais ou distribui\u00e7\u00f5es de Gibbs, modelos cl\u00e1ssicos em processos estoc\u00e1sticos e teoria da mec\u00e2nica estat\u00edstica. Nesta palestra abordarei o problema de estima\u00e7\u00e3o da estrutura de intera\u00e7\u00e3o das vari\u00e1veis (depend\u00eancias condicionais) por meio de um crit\u00e9rio de pseudo-verossimilhan\u00e7a penalizada. Primeiro, introduzimos um crit\u00e9rio para estimar a vizinhan\u00e7a de intera\u00e7\u00e3o de um \u00fanico n\u00f3, que posteriormente ser\u00e1 combinado com as outras vizinhan\u00e7as para obter um estimador do grafo subjacente. Mostrarei resultados de consist\u00eancia do estimador, sem assumir a condi\u00e7\u00e3o de positividade das probabilidades condicionais como \u00e9 usualmente assumido na literatura. Estes resultados abrem possibilidades de estender estes modelos a situa\u00e7\u00f5es de esparsidade, onde muitos par\u00e2metros s\u00e3o nulos. Tamb\u00e9m apresentarei algumas extens\u00f5es em andamento destes resultados para processos satisfazendo condi\u00e7\u00f5es de tipo mixing..<\/p>

25\/06\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Alexandra Mello Schmidt (McGill University, Canad\u00e1)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0A zero-state coupled Markov switching Poisson model for spatio-temporal infectious disease counts.<\/p>

Resumo:<\/strong>\u00a0Spatio-temporal counts of infectious disease cases often contain an excess of zeros. Existing zero inflated Poisson models applied to such data do not adequately capture the switching of the disease between periods of presence and absence overtime. As an alternative, we develop a new zero-state coupled Markov switching Poisson Model, under which the disease switches between periods of presence and absence in each area through a series of partially hidden nonhomogeneous Markov chains coupled between neighboring locations. When the disease is present, an autoregressive Poisson model generates the cases with a possible 0 representing the disease being undetected. Bayesian inference and prediction is illustrated using spatio-temporal counts of dengue fever cases in Rio de Janeiro, Brazil.\u00a0This is joint work with Dirk Douwes-Schultz.<\/p>

18\/06\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Eduardo Fonseca Mendes (EMAp – FGV, Rio de Janeiro)<\/strong><\/p>

T\u00edtulo:<\/strong>\u00a0Sparsity dependent generalized information criteria for regularized m-estimadors..<\/p>

Resumo:<\/strong>\u00a0Resumo: Regularized M-estimators are widely used due to their ability to recover a low-dimensional model in high-dimensional scenarios. Some recent efforts on this subject focused on creating a unified framework for establishing oracle bounds, and deriving conditions for support recovery. Under this same framework, we propose a new Generalized Information Criteria that takes into consideration the sparsity pattern one wishes to recover. We obtain sufficient conditions for model selection consistency of the GIC and path consistency of regularized m-estimators. In other words, we show that under conditions on the penalty function, one may use the GIC for selecting the regularization parameter in a way that the sequence of model subspaces contains the true model with probability converging to one. This allows practical use of the GIC for model selection in high-dimensional scenarios. We illustrate those conditions on examples including LASSO regression and group sparse generalized linear regression.<\/p>

11\/06\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Mariana C\u00fari (ICMC – USP, S\u00e3o Carlos)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Role of deep learning in multidimensional item theory models with correlated latent variables.<\/p>

Resumo:<\/strong>\u00a0 Artificial neural networks with a specific autoencoding structure are capable of estimating parameters for the Multidimensional Logistic 2-Parameter (ML2P) model in Item Response Theory, but with limitations, such as uncorrelated latent traits. In this work, we extend variational autoencoders (VAE) to estimate item parameters and correlated latent abilities, and directly compare the ML2P-VAE method to more traditional parameter estimation methods. In addition, we show that the ML2P-VAE method is capable of estimating parameters for models with high numbers of latent variables with low computational cost, where traditional methods are infeasible for high dimensionality.<\/p>

04\/06\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Wagner Barreto de Souza (KAUST, Ar\u00e1bia Saudita)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Flexible bivariate INGARCH process with a broad range of contemporaneous correlation.<\/p>

Resumo:<\/strong>\u00a0\u00a0We propose a novel flexible bivariate conditional Poisson (BCP) INteger-valued Generalized AutoRegressive Conditional Heteroscedastic (INGARCH) model for correlated count time series data. Our proposed BCP-INGARCH model is mathematically tractable and has as the main advantage over existing bivariate INGARCH models its ability to capture a broad range (both negative and positive) of contemporaneous cross-correlation which is a non-trivial advancement. Properties of stationarity and ergodicity for the BCP-INGARCH process are developed. Estimation of the parameters is performed through conditional maximum likelihood (CML) and finite sample behavior of the estimators are investigated through simulation studies. Asymptotic properties of the \u00a0CML estimators are derived. Additional simulation studies compare and contrast methods of obtaining standard errors of the parameter estimates, where a bootstrap option is demonstrated to be advantageous. Hypothesis testing methods for the presence of contemporaneous correlation between the time series are presented and evaluated. We apply our methodology to monthly counts of hepatitis cases at two nearby Brazilian cities, which are highly cross-correlated. The data analysis demonstrates the importance of considering a bivariate model allowing for a wide range of contemporaneous correlation in real-life applications. Joint work with Luiza S.C. Piancastelli (UCD-Ireland) and Hernando Ombao (KAUST-Saudi Arabia). ArXiv link:\u00a0https:\/\/arxiv.org\/pdf\/2011.08799.pdf<\/a>.<\/p>

28\/05\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Eduardo Guti\u00e9rrez-Pe\u00f1a (Universidad Nacional Aut\u00f3noma de M\u00e9xico)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>General dependence structures for some models based on exponential families with quadratic variance functions.<\/p>

Resumo:<\/strong>\u00a0We describe a procedure to introduce general dependence structures on a set of random variables. These include order-q moving average-type structures, as well as seasonal, periodic and spatial dependencies. The invariant marginal distribution can be in any family that is conjugate to an exponential family with quadratic variance functions. Dependence is induced via latent variables whose conditional distribution mirrors the sampling distribution in a Bayesian conjugate analysis of such exponential families. We obtain strict stationarity as a special case. Joint work with Luis E. Nieto-Barajas, ITAM.<\/p>

ANO DE 2020 – 2\u00ba SEMESTRE<\/strong><\/p>

26\/03\/2021 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Danilo Gilberto de Oliveira Valadares (Doutorando\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Optimal maintenance time for repairable systems.<\/p>

Resumo:<\/strong>\u00a0Um sistema repar\u00e1vel \u00e9 aquele que, quando uma falha ocorre, n\u00e3o \u00e9 descartado e sim restaurado a uma determinada condi\u00e7\u00e3o de opera\u00e7\u00e3o ap\u00f3s um processo de ajuste\/reparo. Neste trabalho, discutiram-se reparos m\u00ednimos ap\u00f3s a falha e reparos preventivos em tempos pr\u00e9-determinados, objetivando encontrar o intervalo de realiza\u00e7\u00e3o do ajuste preventivo que minimize o custo esperado de manuten\u00e7\u00e3o. Quando um reparo m\u00ednimo \u00e9 efetuado, o equipamento volta a funcionar t\u00e3o bom quanto velho e, ap\u00f3s um ajuste preventivo, o equipamento volta a funcionar t\u00e3o bom quanto novo. O processo de falha foi modelado por um Processo Poisson N\u00e3o-Homog\u00eaneo com fun\u00e7\u00e3o intensidade de falhas regida pela Lei das Pot\u00eancias, cujos par\u00e2metros foram estimados utilizando a abordagem cl\u00e1ssica da estat\u00edstica. Os resultados foram exemplificados utilizando um banco de dados real com hist\u00f3rico de falhas em transformadores de energia..<\/p>

26\/03\/2021 \u00e0s 14:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Renata Mendon\u00e7a Rodrigues Vasconcelos (Doutoranda\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo: Risk-adjusted monitoring of time to event in the presence of long-term survivors<\/strong>.<\/p>

Resumo:<\/strong>\u00a0Gr\u00e1ficos de Controle s\u00e3o ferramentas muito \u00fateis em Controle Estat\u00edstico de Processos (CEP) pois auxiliam na detec\u00e7\u00e3o de altera\u00e7\u00f5es na qualidade da produ\u00e7\u00e3o e permitem a investiga\u00e7\u00e3o das poss\u00edveis causas presentes no processo. O CUSUM (cumulative sum) ajustado ao risco surge neste contexto de forma a incorporar o risco espec\u00edfico para cada indiv\u00edduo atrav\u00e9s de estruturas de regress\u00e3o. Nessa perspectiva, considerou-se neste trabalho uma situa\u00e7\u00e3o em que pacientes submetidos a procedimentos m\u00e9dicos podem apresentar diferentes riscos de morte dependendo das diferentes caracter\u00edsticas de cada paciente. Foi proposto ent\u00e3o o uso de um gr\u00e1fico de controle CUSUM ajustado ao risco (RAST CUSUM) para o monitoramento do tempo de vida de pacientes, incorporando no seu processo modelos param\u00e9tricos usuais em sobreviv\u00eancia. No entanto, esses modelos n\u00e3o contemplam a possibilidade de cura de um paciente. O gr\u00e1fico ajustado ao risco RACUF CUSUM foi proposto baseado em um modelo de fra\u00e7\u00e3o de cura como uma extens\u00e3o do RAST CUSUM para o monitoramento de dados de sobreviv\u00eancia com fra\u00e7\u00e3o de cura. Uma ilustra\u00e7\u00e3o da carta de controle proposta foi a partir de dados simulados e com um conjunto de dados reais de pacientes com insufici\u00eancia card\u00edaca atendidos no Instituto do Cora\u00e7\u00e3o (InCor), da Universidade de S\u00e3o Paulo, Brasil..<\/p>

19\/03\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Alvaro Alexander Burbano Moreno (Doutorando\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Hierarchical Bayesian models for predicting spatially correlated curves.<\/p>

Resumo:<\/strong>\u00a0A an\u00e1lise de dados funcionais (FDA) surgiu como uma nova \u00e1rea de investiga\u00e7\u00e3o estat\u00edstica com diversas aplica\u00e7\u00f5es. Na FDA, as unidades s\u00e3o fun\u00e7\u00f5es ou curvas, em que os dados discretos observados s\u00e3o convertidos em fun\u00e7\u00f5es usando v\u00e1rios procedimentos de suaviza\u00e7\u00e3o. Esses dados s\u00e3o ent\u00e3o analisados usando m\u00e9todos estat\u00edsticos tradicionais para extrair informa\u00e7\u00f5es das fun\u00e7\u00f5es. Em certas aplica\u00e7\u00f5es da FDA, a suposi\u00e7\u00e3o de independ\u00eancia condicional \u00e9 razo\u00e1vel; no entanto, esta suposi\u00e7\u00e3o pode n\u00e3o ser v\u00e1lida em configura\u00e7\u00f5es espaciais. Neste artigo os autores apresentam novos modelos Bayesianos baseados em wavelets para dados funcionais espacialmente correlacionados. Estes modelos permitem regularizar as curvas observadas no espa\u00e7o e prever curvas em locais n\u00e3o observados. Compara\u00e7\u00f5es de desempenho s\u00e3o feitas com v\u00e1rias distribui\u00e7\u00f5es a priori para os coeficientes de wavelet e usando um crit\u00e9rio preditivo a posteriori. A proposta \u00e9 ilustrada atrav\u00e9s de dados medindo a porosidade para diversas profundidades de perfura\u00e7\u00f5es no solo.<\/p>

19\/03\/2021 \u00e0s\u00a0 14:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

C\u00e1ssius Henrique Xavier Oliveira (Doutorando\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>A Bayesian joint model of recurrent events and a terminal event.<\/p>

Resumo:\u00a0<\/strong>Recurrent events could be stopped by a terminal event, which commonly occurs in biomedical and clinical studies. In this situation, dependent censoring is encountered because of potential dependence between these two event processes, leading to invalid inference if analyzing recurrent events alone. The joint frailty model is one of the widely used approaches to jointly model these two processes by sharing the same frailty term. One important assumption is that recurrent and terminal event processes are conditionally independent given the subject\u2010level frailty; however, this could be violated when the dependency may also depend on time\u2010varying covariates across recurrences. Furthermore, marginal correlation between two event processes based on traditional frailty modeling has no closed form solution for estimation with vague interpretation. In order to fill these gaps, we propose a novel joint frailty\u2010copula approach to model recurrent events and a terminal event with relaxed assumptions. Metropolis\u2013Hastings within the Gibbs Sampler algorithm is used for parameter estimation. Extensive simulation studies are conducted to evaluate the efficiency, robustness, and predictive performance of our proposal. The simulation results show that compared with the joint frailty model, the bias and mean squared error of the proposal is smaller when the conditional independence assumption is violated. Finally, we apply our method into a real example extracted from the MarketScan database to study the association between recurrent strokes and mortality.<\/p>

12\/03\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Leonardo Angelo Soares da Silva (Doutorando\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Uma nova cota para o n\u00famero crom\u00e1tico ac\u00edclico de arestas.<\/p>

Resumo:\u00a0<\/strong>Resumo: Nesta apresenta\u00e7\u00e3o, ser\u00e1 exposta uma nova cota que foi obtida para o n\u00famero crom\u00e1tico de aresta ac\u00edclica, a'(G), de um grafo G com grau m\u00e1ximo \u0394 mostrando que tal \u00edndice \u00e9 de a'(G) \u2264 3,569(\u0394 \u2212 1). Para isso, partiremos do princ\u00edpio de uma an\u00e1lise probabil\u00edstica de um algoritmo semelhante, realizado anteriormente por Giotis et al. que obteve a cota a'(G) \u2264 3,74(\u0394 \u2212 1). Desse modo, os autores revisaram e modificaram ligeiramente o m\u00e9todo descrito por Giotis obtendo, com isso, uma melhora no \u00edndice crom\u00e1tico a'(G).<\/p>

05\/03\/2021 \u00e0s\u00a0 14:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Larissa Natany Almeida Martins (Doutoranda\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>A Bayesian network approach for population synthesis.<\/p>

Resumo:\u00a0<\/strong>Agent-based micro-simulation models require a complete list of agents with detailed demographic\/socioeconomic information for the purpose of behavior modeling and simulation. This paper introduces a new alternative for population synthesis based on Bayesian networks. A Bayesian network is a graphical representation of a joint probability distribution, encoding probabilistic relationships among a set of variables in an efficient way. Similar to the previously developed probabilistic approach, in this paper, we consider the population synthesis problem to be the inference of a joint probability distribution. In this sense, the Bayesian network model becomes an efficient tool that allows us to compactly represent\/reproduce the structure of the population system and preserve privacy and confidentiality in the meanwhile. We demonstrate and assess the performance of this approach in generating synthetic population for Singapore, by using the Household Interview Travel Survey (HITS) data as the known test population. Our results show that the introduced Bayesian network approach is powerful in characterizing the underlying joint distribution, and meanwhile the overfitting of data can be avoided as much as possible.<\/p>

05\/03\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Marcio Augusto Ferreira Rodrigues (Doutorando\u00a0– DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Semiparametric regression analysis of interval-censored competing risks data.<\/p>

Resumo:\u00a0<\/strong>Interval-censored competing risks data arise when each study subject may experience an event or failure from one of several causes and the failure time is not observed directly but rather is known to lie in an interval between two examinations. We formulate the effects of possibly time-varying (external) covariates on the cumulative incidence or sub-distribution function of competing risks (i.e., the marginal probability of failure from a specific cause) through a broad class of semiparametric regression models that captures both proportional and non-proportional hazards structures for the sub-distribution. We allow each subject to have an arbitrary number of examinations and accommodate missing information on the cause of failure. We consider nonparametric maximum likelihood estimation and devise a fast and stable EM-type algorithm for its computation. We then establish the consistency, asymptotic normality, and semiparametric efficiency of the resulting estimators for the regression parameters by appealing to modern empirical process theory. In addition, we show through extensive simulation studies that the proposed methods perform well in realistic situations. Finally, we provide an application to a study on HIV-1 infection with different viral subtypes.<\/p>

26\/02\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Ana J\u00falia Alves C\u00e2mara (Doutoranda – DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>On generalized additive models with dependent time series covariate.<\/p>

Resumo:\u00a0<\/strong>The generalized additive model (GAM) is a standard statistical methodology and is frequently used in various fields of applied data analysis where the response variable is non-normal, e.g., integer valued, and the explanatory variables are continuous, typically normally distributed. Standard assumptions of this model, among others, are that the explanatory variables are independent and identically distributed vectors which are not multicollinear. To handle the multicollinearity and serial dependence together a new hybrid model, called GAM-PCA-VAR model, was proposed in [17] which is the combination of GAM with the principal component analysis (PCA) and the vector autoregressive (VAR) model. In this paper, some properties of the GAM-PCA-VAR model are discussed theoretically and verified by simulation. A real data
set is also analysed with the aim to describe the association between respiratory disease and air pollution concentrations.<\/p>

19\/02\/2021 \u00e0s\u00a0 14:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Marta Cristina Colozza Bianchi (Doutoranda – DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Modelos de mistura com depend\u00eancia Markoviana para identificar observa\u00e7\u00f5es at\u00edpicas em s\u00e9rie temporal com espa\u00e7amento irregular.<\/strong><\/p>

Resumo:\u00a0<\/strong>Neste semin\u00e1rio ser\u00e3o apresentados dois modelos Bayesianos de mistura com depend\u00eancia Markoviana. A modelagem \u00e9 motivada por duas aplica\u00e7\u00f5es para an\u00e1lise de milhares de medi\u00e7\u00f5es de express\u00e3o g\u00eanica, em tumores de alguns tipos de c\u00e2ncer, cujas localiza\u00e7\u00f5es s\u00e3o mapeadas em cromossomos definindo s\u00e9ries com espa\u00e7amento irregular. Este tipo de modelo foi proposto em Mayrink e Gon\u00e7alves (2017) com aplica\u00e7\u00e3o a dados de microarray, e estendido em Mayrink e Gon\u00e7alves (2020) com aplica\u00e7\u00e3o a dados RNA-Seq. Em ambos os estudos, o objetivo \u00e9 identificar observa\u00e7\u00f5es at\u00edpicas. No contexto de microarrays, deseja-se detectar regi\u00f5es gen\u00f4micas associadas a valores de alta express\u00e3o (superexpress\u00e3o), que definem clusters de observa\u00e7\u00f5es consecutivas. J\u00e1 na an\u00e1lise de RNA-Seq, o objetivo \u00e9 encontrar dois tipos de regi\u00f5es cromoss\u00f4micas: superexpress\u00e3o e subexpress\u00e3o. As caracter\u00edsticas de alta ou baixa express\u00e3o g\u00eanica s\u00e3o importantes para estudar a progress\u00e3o de um c\u00e2ncer. Atrav\u00e9s delas identificam-se regi\u00f5es contendo genes com atividade diferenciada na doen\u00e7a. Em Mayrink e Gon\u00e7alves (2017) o modelo desenvolvido considera uma mistura de distribui\u00e7\u00f5es com m\u00e9dias ordenadas de forma que o \u00faltimo componente seja respons\u00e1vel por acomodar genes superexpressos. No trabalho de 2020, o primeiro e \u00faltimo componentes da mistura incorporam os genes subexpressos e superexpressos, respectivamente. O modelo \u00e9 flex\u00edvel o suficiente para lidar de forma eficiente com o espa\u00e7amento irregular dos dados ao usar as informa\u00e7\u00f5es de dist\u00e2ncia entre express\u00f5es vizinhas para inferir sobre a exist\u00eancia de uma depend\u00eancia Markoviana. Esta depend\u00eancia tem papel chave para a detec\u00e7\u00e3o das regi\u00f5es de interesse. A infer\u00eancia Bayesiana \u00e9 realizada por meio de amostragem indireta via algoritmos MCMC.<\/p>

19\/02\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Magda Carvalho Pires – DEST\/UFMG<\/strong>\u00a0(Joint work with Milena S. Marcolino, Lucas E. F. Ramos, Rafael T. Silva, Luana M. Oliveira et.al)<\/p>

T\u00edtulo:\u00a0<\/strong>ABC2<\/sub><\/span>-SPH risk score for in-hospital mortality in COVID-19 patients: development, external validation and comparison with other available scores.<\/strong><\/p>

Resumo:\u00a0<\/strong>Coronavirus disease 2019 (COVID-19), caused by the SARS-CoV-2 virus, is still the main global health, social and economic challenge. In this context, fast and efficient assessment of prognosis of the disease is needed to optimize the allocation of health care and human resources, to empower early identification and intervention of patients at higher risk of poor outcome. Thus, rapid scoring systems, which combine different variables to estimate the risk of poor outcome, may be extremely helpful for fast and effective assessment of those patients in the emergency department. Following international guidelines, generalized additive models and LASSO logistic regression were performed to develop a prediction model for in-hospital mortality, based on the 3978 patients that were admitted during March-July, 2020. The model was validated in the 1054 patients admitted during August-September 30, as well as in an external cohort of 474 Spanish patients. Our ABC2-SPH score showed good discrimination, calibration and overall performance in both Brazilian cohorts, but, in the Spanish cohort, mortality was somewhat underestimated in patients with very high (>25%) risk. The ABC2-SPH score is implemented in a freely available online risk calculator (<\/span>https:\/\/abc2sph.com\/<\/a>).<\/span><\/p>

Preprint:\u00a0www.medrxiv.org\/content\/10.1101\/2021.02.01.21250306v1<\/a><\/p>

05\/02\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Joseph Lucas (Senior Research Scientist na Caravan Health, EUA)<\/strong><\/p>

T\u00edtulo: <\/strong>A practical guide to prediction using temporal event data.<\/p>

Resumo:\u00a0<\/strong>We look at some practical aspects to prediction using (potentially high dimensional) temporal event data. The talk will touch on (i) feature extraction,\u00a0(ii) overfitting, (iii) using a model agnostic approach, (iv) variable importance, and (v) managing computing resources. We will demonstrate the\u00a0techniques by building models to predict device failures from connected monitors (low dimensional) and to predict end of life events from medical\u00a0records and claims data (high dimensional).<\/span><\/p>

29\/01\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>\u00a0<\/strong><\/p>

Fabrizio Ruggeri (CNR IMATI,Italy)<\/strong><\/p>

T\u00edtulo: <\/strong>Likelihood-Free Parameter Estimation for Dynamic Queueing Networks\u00a0The case of the immigration queue at an international airport.<\/p>

Resumo:<\/strong><\/p>

Many complex real-world systems such as airport terminals, manufacturing\u00a0processes and hospitals are modelled with networks of queues. To estimate\u00a0parameters, restrictive assumptions are usually placed on these models. For\u00a0instance arrival and service distributions are assumed to be time-invariant.\u00a0Violating this assumption are so-called dynamic queueing networks (DQNs)\u00a0which are more realistic but do not allow for likelihood-based parameter\u00a0estimation. We consider the problem of using data to estimate the parameters of a DQN.\u00a0The is the first example of Approximate Bayesian Computation (ABC) being\u00a0used for parameter inference of DQNs. We combine computationally efficient\u00a0simulation of DQNs with ABC and an estimator for maximum mean discrepancy.\u00a0DQNs are simulated in a computationally efficient manner with Queue Departure\u00a0Computation (a simulation techniques we are proposing), without the need for\u00a0time-invariance assumptions, and parameters are inferred from data without\u00a0strict data-collection requirements. Forecasts are made which account for\u00a0parameter uncertainty. We embed this queueing simulation within an ABC\u00a0sampler to estimate parameters for DQNs in a straightforward manner. We\u00a0motivate and demonstrate this work with the example of passengers arriving at\u00a0the passport control in an international airport.<\/p>

Joint work with Anthony Ebert, Kerrie Mengersen, Paul Wu, Antonietta Mira\u00a0and Ritabrata Dutta. Available: https:\/\/arxiv.org\/abs\/1804.02526<\/p>

22\/01\/2021 \u00e0s\u00a0 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Francisco Cribari-Neto (Departamento de Estat\u00edstica-UFPE)<\/strong><\/p>

T\u00edtulo: <\/strong>Improved testing inferences for beta regressions with parametric mean link function.<\/p>

Resumo:<\/strong><\/p>

Beta regressions are widely used for modeling random variables that assume values in the standard unit interval, (0,1), such as rates, proportions, and income concentration indices. Parameter estimation is typically performed via maximum likelihood and hypothesis testing inferences on the model parameters are commonly performed using the likelihood ratio test. Such a test, however, may deliver inaccurate inferences when the sample size is small. It is thus important to develop alternative testing procedures that are more accurate when the sample contains only few observations. In this paper, we introduce the beta regression model with parametric mean link function and derive two modified likelihood ratio test statistics for that class of models. We provide simulation evidence that shows that the new tests usually outperform the standard likelihood ratio test in samples of small to moderate sizes. We also present and discuss two empirical applications.<\/p>

15\/01\/2021 \u00e0s\u00a014:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Manuel Galea (Pontificia Universidad Catolica de Chile)<\/strong><\/p>

T\u00edtulo: <\/strong>Robust inference in the Capital Assets Pricing Model using the multivariate t\u2212distribution.<\/p>

Resumo:<\/strong><\/p>

In this work we consider the Capital Asset Pricing Model under the multivariate t\u2212distribution with finite second moment. This distribution, which contain the normal distribution, offer a more flexible framework for modeling asset returns. The main objective is to develop statistical inference tools, such as parameter estimation and linear hypothesis tests in asset pricing models, with an emphasis on the Capital Asset Pricing Model (CAPM). An extension of the CAPM, the Multifactor Asset Pricing Model (MAPM), is also discussed. A simple algorithm to estimate the model parameters, including the kurtosis parameter, is implemented. Analytical expressions for the Score function and Fisher information matrix are provided. For linear hypothesis tests, the four most widely used tests (likelihood-ratio, Wald, score, and gradient statistics) are considered. In order to test the mean-variance efficiency, explicit expressions for these four statistical tests are also presented. The results are illustrated using two real data sets: the Chilean Stock Market data set and another from the New York Stock Exchange. The asset pricing model under the multivariate t-distribution presents a good fit, clearly better than the asset pricing model under the assumption of normality, in both data sets.<\/p>

08\/01\/2021 \u00e0s\u00a013:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Ivair Silva (UFOP)<\/strong><\/p>

T\u00edtulo: <\/strong>Fixed-Length Confidence Intervals Following a Sequential Test with Binomial Data.<\/p>

Resumo:<\/strong><\/p>

Sample size and time to detect a signal are key performance measures in statistical sequential hypothesis testing. While the former should be optimized in Phase III clinical trials, minimizing the latter is of major importance in post-marked drug and vaccine safety surveillance of adverse events. However, in practice, even when strong evidences indicate that the surveillance could be stopped for drawing a test decision, it may be necessary to continue collecting data in order to improve the precision of the point estimator. For binomial data, this paper presents a linear programming framework to find the optimal alpha spending that provides fixed-width and fixed-accuracy confidence intervals for the relative risk parameter. The solution permits minimization of expected time to signal, or expected sample size as needed. In addition, the method is extended for group sequential testing with variable Bernoulli probabilities. To illustrate, we use simulated data mimicking actual clinical trials on experimental COVID-19 treatments.Fixed-Length Confidence Intervals Following a Sequential Test with Binomial Data.<\/p>

ANO DE 2020 – 1\u00ba SEMESTRE<\/strong><\/p>

11\/12\/2020 \u00e0s\u00a013:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Marcos Prates (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Spatial Confounding Beyond Generalized Linear Mixed Models: Extension to Shared Components and Spatial Frailty Models.<\/p>

Resumo:<\/strong><\/p>

Spatial confounding is defined as the confounding between the fixed and spatial random effects in generalized linear mixed models (GLMMs). It gained attention in the past years, as it may generate unexpected results in modeling. We introduce solutions to alleviate the spatial confounding beyond GLMMs for two families of statistical models. In the shared component models, multiple count responses are recorded at each spatial location, which may exhibit similar spatial patterns. Therefore, the spatial effect terms may be shared between the outcomes in addition to specifics spatial patterns. Our proposal relies on the use of modified spatial structures for each shared component and specific effects. Spatial frailty models can incorporate spatially structured effects and it is common to observe more than one sample unit per area which means that the support of fixed and spatial effects differs. Thus, we introduce a projection-based approach for reducing the dimension of the data. An R package named “RASCO: An R package to Alleviate Spatial Confounding” is provided. Cases of lung and bronchus cancer in the state of California are investigated under both methodologies and the results prove the efficiency of the proposed methodology..<\/p>

04\/12\/2020 \u00e0s\u00a013:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Guido Moreira (P\u00f3s-Doc, DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/b>Analysis of presence-only data via exact Bayes, with model and effects identification.<\/p>

Resumo:\u00a0<\/strong>This paper provides an exact modeling approach for the analysis of presence-only ecological data. The approach is also based on frequently used Inhomogeneous Poisson Process but unlike other approaches does not rely on model approximations for performing inference. Exactness is achieved via a data augmentation scheme. One of the augmented processes can be interpreted as the unobserved occurrences of the relevant species and its posterior distribution can be used to make predictions of the species over the region of study beyond the observer bias. The data augmentation also provides a natural Gibbs sampler to make Bayesian inference through MCMC. The proposal shows better AUC prediction metric than the traditional Poisson Process whose intensity function is log-linear with respect to the covariates, which is currently the standard method. Additionally, an identifiability problem that arises in the traditional model does not affect our proposal and this is verified on analyses with real ecological data.<\/span><\/p>

06\/11\/2020 \u00e0s\u00a013:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Murray Pollock (Newcastle University, UK)<\/strong><\/p>

T\u00edtulo: <\/strong>The Restore Process – Practical CFTP by enriching Markov processes.<\/p>

Resumo:\u00a0<\/strong>We develop a new class of Markov processes comprising local dynamics governed by a fixed Markov process, which are enriched with regenerations from a fixed distribution at a state-dependent rate. We give conditions under which such processes possess a given target distribution as their invariant measures, thus making them amenable for use within Monte Carlo methodologies. Enrichment imparts a number of desirable theoretical and methodological properties, which includes straightforward conditions for the process to be uniformly ergodic and possess a coupling from the past construction that enables exact sampling from the invariant distribution. Joint work with David Steinsaltz \/ Gareth Roberts \/ Andi Wang..<\/span><\/p>

30\/10\/2020 \u00e0s\u00a010:00 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Leonardo Brand\u00e3o (UFMG-Semin\u00e1rios 1B)<\/strong><\/p>

T\u00edtulo: <\/strong>The poly-logWeibull model applied to space-time interpolation of temperature.<\/p>

Resumo:\u00a0<\/strong>In this paper, a multivariate log-Weibull model for spatially dependent data is\u00a0defined by marginalizing a conditional Pareto distribution with respect to a\u00a0shared spatial random effect of alpha-stable distributions. Some properties of\u00a0this newmodel are derived, and procedures for the estimation and inference are\u00a0discussed. An application is developed to study observed temperature data sets\u00a0collected from weather stations in the Brazilian Amazon.<\/span><\/p>

Paper by A. L. Mota, M. S. De Lima , F. N. Demarqui e L. H. Duczmal<\/p>

23\/10\/2020 \u00e0s\u00a014:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Rams\u00e9s H. Mena (IIMAS-UNAM, Mexico)<\/strong><\/p>

T\u00edtulo: <\/strong>Beta-binomial stick-breaking non-parametric prior.<\/p>

Resumo:\u00a0<\/strong>A new class of nonparametric prior distributions, termed Beta-Binomial stick-breaking process, is proposed. By allowing the underlying length random variables to be dependent through a Beta marginals Markov chain, an appealing discrete random probability measure arises. The chain\u2019s dependence parameter controls the ordering of the stick-breaking weights, and thus tunes the model\u2019s label-switching ability. Also, by tuning this parameter, the resulting class contains the Dirichlet process and the Geometric process priors as particular cases, which is of interest for MCMC implementations.<\/span><\/p>

Some properties of the model are discussed and a density estimation algorithm is proposed and tested with simulated datasets.<\/p>

Reference: Gil-Leyva, M.F., Mena, R.H. and Nicoleris, T. (2020). Beta-Binomial stick-breaking non-parametric prior Electronic Journal of Statistics. 14, 1479-1507. https:\/\/doi.org\/10.1214\/20-EJS1694.<\/p>

16\/10\/2020 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Raffaele Argiento (Department of Statistics, Universit\u00e0 Cattolica del Sacro Cuore)<\/strong><\/p>

T\u00edtulo: <\/strong>Is Infinity That Far? A Bayesian Nonparametric Perspective of Finite Mixture Models.<\/p>

Resumo:\u00a0<\/strong>Mixture models are one of the most widely used statistical tools when dealing with data from heterogeneous populations. This talk considers the long-standing debate over finite mixture and infinite mixtures and brings the two modelling strategies together, by showing that a finite mixture is simply a realization of a point process. Following a Bayesian nonparametric perspective, we introduce a new class of prior: the Normalized Independent Point Processes. We investigate the probabilistic properties of this new class. Moreover, we design a conditional algorithm for finite mixture models with a random number of components overcoming the challenges associated with the Reversible Jump scheme and the recently proposed marginal algorithms. We illustrate our model on real data and discuss a relevant application in population genetics.<\/span><\/p>

09\/10\/2020 \u00e0s 14:00 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Peter M\u00fcller (University of Texas)<\/strong><\/p>

T\u00edtulo: <\/strong>Bayesian Categorical Matrix Factorization via Double Feature Allocation.<\/p>

Resumo:\u00a0<\/strong>We propose a categorical matrix factorization method to infer latent diseases from electronic health records data. A latent disease is defined as an unknown cause that induces a set of common symptoms for a group of patients. The proposed approach is based on a novel double feature allocation model which simultaneously allocates features to the rows and the columns of a categorical matrix. Using a Bayesian approach, available prior information on known diseases greatly improves identifiability of latent diseases. This includes known diagnoses for patients and known association of diseases with symptoms. For application to large data sets, as they naturally arise in electronic health records, we develop a divide-and-conquer Monte Carlo algorithm, which allows inference for the proposed double feature allocation model, and a wide range of related Bayesian nonparametric mixture models and random subsets. We validate the proposed approach by simulation studies including mis-specified models and comparison with sparse latent factor models. In an application to Chinese electronic health records (EHR) data, we find results that agree with related clinical and medical knowledge.<\/span><\/p>

References:<\/b><\/p>

1) Bayesian Double Feature Allocation for Phenotyping with Electronic Health Records, Yang Ni, Peter Mueller, Yuan Ji<\/p>

https:\/\/arxiv.org\/abs\/1809.08988<\/a><\/p>

Journal of the American Statistical Association, in press.<\/p>

2) Consensus Monte Carlo for Random Subsets using Shared Anchors, Yang Ni, Yuan Ji, and Peter Mueller<\/p>

https:\/\/arxiv.org\/abs\/1906.12309<\/a><\/p>

Journal of Computational and Graphical Statistics}, in press.<\/p>

02\/10\/2020 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Alexandre Galv\u00e3o Patriota (USP)<\/strong><\/p>

T\u00edtulo: <\/strong>Modelos de regress\u00e3o el\u00edpticos com parametriza\u00e7\u00e3o geral.<\/p>

Resumo:<\/strong>\u00a0Neste semin\u00e1rio irei apresentar alguns dos resultados assint\u00f3ticos desenvolvidos considerando modelos\u00a0de regress\u00e3o el\u00edpticos com parametriza\u00e7\u00e3o geral. Estes modelos incluem modelos mistos, modelos n\u00e3o lineares heterosced\u00e1sticos, modelos com erros nas vari\u00e1veis, entre outros.<\/p>

25\/09\/2020 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Kelly Cristina Mota Gon\u00e7alves (DME-UFRJ)<\/strong><\/p>

T\u00edtulo: <\/strong>Bayesian dynamic quantile linear models and some extensions.<\/p>

Resumo:<\/strong>\u00a0The main aim of this talk is to present a new class of models, named dynamic quantile linear models. It combines dynamic linear models with distribution free quantile regression producing a robust statistical method. This class of models provides richer information on the effects of the predictors than does the traditional mean regression and it is very insensitive to heteroscedasticity and outliers, accommodating the non-normal errors often encountered in practicalapplications. Bayesian inference for quantile regression proceeds by forming the likelihood function based on the asymmetric Laplace distribution and a location-scale mixture representation of it allows finding analytical expressions for the conditional posterior densities of the model. Thus, Bayesian inference for dynamic quantile linear models can be performed using an efficient Markov chain Monte Carlo algorithm or a fast sequential procedure suited for high-dimensional predictive modeling applications with massive data. Finally, a hierarchical extension, useful to account for structural features in the dataset, will be also presented.<\/p>

18\/09\/2020 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Denise Duarte (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Modelos de redes de afinidade<\/p>

Resumo:<\/strong>\u00a0Uma das abordagens mais populares atualmente na literatura sobre dados relacionais \u00e9 a An\u00e1lise de Redes Complexas. Consequentemente, an\u00e1lises estat\u00edsticas sobre redes sociais buscaram acompanhar este crescimento para atender \u00e0 esta demanda. Para modelar estatisticamente os fen\u00f4menos estudados em redes socais, modelos probabil\u00edsticos em grafos aleat\u00f3rios tem sido bastante utilizados. Entretanto, as redes sociais possuem caracter\u00edsticas que s\u00e3o diferentes dos modelos de grafos aleat\u00f3rios que possuem arestas independentes. A proposta deste trabalho \u00e9 apresentar e estudar um modelo de grafo aleat\u00f3rio onde as liga\u00e7\u00f5es (arestas) s\u00e3o baseadas nas caracter\u00edsticas dos v\u00e9rtices, permitindo uma modelagem mais realista de uma rede. Propomos uma vasta fam\u00edlia de modelos, que chamamos de Modelos de Redes de Afinidade, onde as conex\u00f5es s\u00e3o valoradas segundo uma fun\u00e7\u00e3o que mensura a afinidade entre os atores da rede. Al\u00e9m disso, as conex\u00f5es s\u00e3o realizadas a partir de um determinado ponto de corte para o valor desta fun\u00e7\u00e3o afinidade, de acordo com o n\u00edvel de afinidade desejado pelo pesquisador. Para exemplificar o estudo do comportamento do nosso modelo, elaboramos um estudo simulado baseado em simula\u00e7\u00f5es de Monte Carlo para uma das fun\u00e7\u00f5es de afinidade descritas neste trabalho. Realizamos uma calibra\u00e7\u00e3o nos par\u00e2metros geradores do modelo, analisando suas medidas topol\u00f3gicas, comparando com as medidas topol\u00f3gicas encontradas em grafos com a mesma distribui\u00e7\u00e3o de afinidade, mas com arestas sorteadas independentemente. O estudo mostra que o Modelo de Redes de Afinidade consegue capturar caracter\u00edsticas importantes de redes sociais. Trabalho em conjunto com Wesley H.S. Pereira e Rodrigo B. Ribeiro.<\/p>

11\/09\/2020 \u00e0s 13:30 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Fl\u00e1vio Bambirra Gon\u00e7alves (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Exact and computationally efficient Bayesian inference for generalized Markov modulated Poisson processes<\/p>

Resumo:<\/strong>\u00a0Statistical modeling of point patterns is an important and common problem in several areas. The Poisson process is the most common process used for this purpose, in particular, its generalization that considers the intensity function to be stochastic. This is called a Cox process and different choices to model the dynamics of the intensity gives raise to a wide range of models. We present a new class of unidimensional Cox process models in which the intensity function assumes parametric functional forms that switch among them according to a continuous-time Markov chain. A novel methodology is proposed to perform exact Bayesian inference based on MCMC algorithms. The term exact refers to the fact that no discrete time approximation is used and Monte Carlo error is the only source of inaccuracy. The reliability of the algorithms depends on a variety of specifications which are carefully addressed, resulting in a computationally efficient (in terms of computing time) algorithm and enabling its use with large datasets. Simulated and real examples are presented to illustrate the efficiency and applicability of the proposed methodology. A specific model to fit epidemic curves is proposed and used to analyze data from Dengue Fever in Brazil and COVID-19 in some countries.This is joint work with Livia Dutra and Roger Silva.<\/p>

04\/09\/2020 \u00e0s 13:00 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Hedibert Freitas Lopes (Insper-SP)<\/strong><\/p>

T\u00edtulo: <\/strong>The Illusion of the Illusion of Sparsity<\/p>

Resumo:<\/strong>\u00a0The emergence of Big Data raises the question of how to model statistical series when there is a large number of possible regressors. This article addresses the issue by comparing the possibility of using dense or sparse models in a Bayesian approach, allowing for variable selection and shrinkage. We discuss the results reached by Giannone, Lenza, and Primiceri (2018) through a \u201cSpike-and-Slab\u201d prior, which suggest an \u201cillusion of sparsity\u201d in economic datasets, as no clear patterns of sparsity could be found. We make a further revision of the posterior distributions of the model, and propose three experiments to evaluate the robustness of the adopted prior distribution. We find that the model indirectly induces variable selection and shrinkage, what suggests that the \u201cillusion of sparsity\u201d is, itself, an illusion. Note: Joint work with Bruno Vinicius Nunes Fava and was part of his 2019 undergraduate final projection Economics at Insper. Bruno starts his PhD in Economics at Northwestern University in August 2020.<\/p>

28\/08\/2020 \u00e0s 11:00 hs\u00a0–\u00a0Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong><\/p>

Oliver Stone and Theo Economou\u00a0(Institute for Data Science and Artificial IntelligenceUniversity of Exeter)<\/strong><\/p>

T\u00edtulo: <\/strong>Bayesian Hierarchical Frameworks for Correcting Under-reporting and Delayed Reporting of Count Data<\/p>

Resumo:<\/strong>\u00a0The Covid-19 pandemic has brought renewed attention on the limitations of systems which report cases and deaths, specifically under-reporting and delayed reporting. In this two-part seminar, we will discuss Bayesian hierarchical approaches to correcting these issues, to enable enhanced monitoring and decision-making. Finally, we will demonstrate how the framework for correcting delayed reporting can be used for now-casting and forecasting of English hospital deaths from Covid-19.<\/p>

21\/08\/2020 \u00e0s 13:30hs – Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>
<\/strong><\/p>

Rafael Bassi Stern (UFSCar)<\/strong><\/p>

T\u00edtulo: <\/strong>CD-Split: Efficient Conformal Regions in High Dimensions<\/p>

14\/08\/2020 \u00e0s 13:30h – Local:\u00a0Canal do Youtube: Semin\u00e1rios DEST – UFMG<\/a><\/strong>
<\/strong><\/p>

Luiz Max Carvalho (EMAP-FGV)<\/strong><\/p>

T\u00edtulo: <\/strong>Adaptive Markov chain Monte Carlo on the space of time-calibrated trees<\/p>

20\/03\/2020 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Luiz Max Carvalho (EMAP-FGV)<\/strong><\/p>

T\u00edtulo: <\/strong>Efficient transition kernels for Bayesian phylogenetics<\/p>

13\/03\/2020 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Vinicius Mayrink (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Structural equation modeling with time dependence: an application comparing Brazilian energy distributors<\/p>

ANO DE 2019 – 2\u00ba SEMESTRE<\/strong><\/p>

06\/12\/2019 \u00e0s 14:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Walmir dos Reis Miranda Filho (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Frailty and Copula Models: Similarities and Differences<\/p>

04\/12\/2019 \u00e0s 14:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Daiane Zuanetti (UFSCar)
<\/strong><\/p>

T\u00edtulo: <\/strong>Subset nonparametric Bayesian clustering – an application in genetic data<\/p>

04\/12\/2019 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Rafael Izbicki (UFSCar)
<\/strong><\/p>

T\u00edtulo: <\/strong>Quantification under prior probability shift: the ratio estimator and its extensions<\/p>

29\/11\/2019 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Edson Ferreira (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Context Tree Estimation for Not Necessarily Finite Memory Processes, Via BIC and MDL<\/p>

22\/11\/2019 \u00e0s 14:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Renan Xavier Cortes\u00a0(Anglo American)\u00a0<\/strong><\/p>

T\u00edtulo: <\/strong>Building open-source tools in Python for Spatio-Temporal Data and Modelling<\/p>

22\/11\/2019 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Patricia Viana (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Bayesian Cluster Analysis: Point Estimation and Credible Balls<\/p>

08\/11\/2019 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Jussiane Gon\u00e7alves (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Zero-inflated mixed Poisson regression models<\/p>

01\/11\/2019 \u00e0s 14:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Diogo Carlos dos Santos (UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>O processo de percola\u00e7\u00e3o de grau restrito<\/p>

01\/11\/2019 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Guilherme Ludwig (UNICAMP)<\/strong><\/p>

T\u00edtulo: <\/strong>Interacting cluster point process model for epidermal nerve fiberss<\/p>

25\/10\/2019 \u00e0s 13:30h – Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Glaura C. Franco (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Non-Gaussian Time Series Models<\/p>

18\/10\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Ronald Dickman (F\u00edsica-UFMG)<\/strong><\/p>

T\u00edtulo:<\/strong> Steady-state thermodynamics and phase coexistence far from equilibrium<\/p>

11\/10\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Dani Gamerman (DEST e UFRJ))<\/strong><\/p>

T\u00edtulo: <\/strong>Modelagem hier\u00e1rquica em problemas de alta dimens\u00e3o.<\/p>

04\/10\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Fabricio Murai (DCC-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Reasoning from Partially Observed Networks: Sampling, Estimation and Models.<\/p>

27\/09\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Ilka Afonso Reis (DEST\/UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Um breve passeio pela Psicometria: minha experi\u00eancia com valida\u00e7\u00e3o de instrumentos.<\/p>

20\/09\/2019 \u00e0s 13:00h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Vera Tomazella (UFSCar)<\/strong><\/p>

T\u00edtulo: <\/strong>Defective Models Induced By Gamma Frailty Term for Survival Data With Cured Fraction<\/p>

13\/09\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Roberto Nalon (DCC- Big Data)<\/strong><\/p>

T\u00edtulo: <\/strong>Detecting Spatial Clusters of Disease Infection Risk Using Sparsely Sampled Social Media Mobility Patterns<\/p>

11\/09\/2019 \u00e0s 11:00h –\u00a0 Local: LCC – ICEx<\/strong>\u00a0<\/strong><\/p>

Ian M Danilevicz (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>An overview of robust spectral estimators<\/p>

06\/09\/2019 \u00e0s 13:30h –\u00a0 Local: Audit\u00f3rio III do – ICEx<\/strong>\u00a0<\/strong><\/p>

SEMIN\u00c1RIO CI\u00caNCIA DE DADOS: McKinsey & Company<\/strong><\/p>

Na sexta-feira dia 06 de setembro, o Departamento de Estat\u00edstica estar\u00e1 recebendo, no Audit\u00f3rio III do ICEx, o Data Scientist, Marcus Watari, e o Data Engineer, Daniel Golhiardi, ambos consultores da McKinsey & Company. Eles apresentar\u00e3o um Semin\u00e1rio para alunos da P\u00f3s-gradua\u00e7\u00e3o em Estat\u00edstica, Qu\u00edmica, F\u00edsica, Ci\u00eancia da Computa\u00e7\u00e3o, Matem\u00e1tica e Engenharia El\u00e9trica. Mostrar\u00e3o casos de como a ci\u00eancia de dados tem sido aplicada em contextos reais em diferentes ind\u00fastrias e quais s\u00e3o os desafios e possibilidades de atua\u00e7\u00e3o na carreira de um Engenheiro e Cientista de Dados.<\/p>

\"M2\"<\/p>

\u00a023\/08\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Marcelo Azevedo Costa (Eng. Produ\u00e7\u00e3o \u2013 UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Failure detection in robotic arms using statistical modeling, machine learning and hybrid gradient boosting<\/p>

09\/08\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Michel Spira (Departamento de Matem\u00e1tica – UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Matem\u00e1tica e o Homem Vitruviano<\/p>

01\/08\/2019 \u00e0s 14:00h –\u00a0 Local: sala 3060 – ICEx<\/strong>\u00a0<\/strong><\/p>

Alexandre Gaudilli\u00e8re (CNRS-Marseille); Joint work: A. Bianchi (Universita di Padova); P. Milanesi (Universite d’Aix-Marseille); M. E. Vares(UFRJ)<\/strong><\/p>

T\u00edtulo: <\/b>Exponential transition law for the kinetic Ising model.<\/p>

ANO DE 2019 – 1\u00ba SEMESTRE<\/strong><\/p>

05\/07\/2019 \u00e0s 14:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Fernanda Gabriely Batista Mendes (DEST)
<\/strong><\/p>

SEMIN\u00c1RIO 2 – T\u00edtulo: <\/b>Constru\u00e7\u00e3o de cadeia de Markov estacion\u00e1ria.<\/p>

05\/07\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Adrian Luna (DEST)
<\/strong><\/p>

SEMIN\u00c1RIO 1 – T\u00edtulo: <\/b>Redes Aleat\u00f3rias: alguns desafios.<\/p>

28\/06\/2019 \u00e0s 15:00h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Hernando Ombao – <\/strong>Biostatistics Research Group – STAT Program – King Abdullah University of Science and Technology (KAUST, Saudi Arabia)<\/b><\/p>

T\u00edtulo: <\/strong>Exploring the Dependence Structure in Multivariate Time Series.<\/p>

28\/06\/2019 \u00e0s 14:00h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Hernando Ombao – Biostatistics Research Group – STAT Program – King Abdullah University of Science and Technology (KAUST, Saudi Arabia)
<\/strong><\/p>

T\u00edtulo: <\/strong>Spectral and Coherence Analysis: Basic Ideas and Applications.<\/p>

07\/06\/2019 \u00e0s 13:30h –\u00a0 Local: sala 2076 – ICEx<\/strong>\u00a0<\/strong><\/p>

Michelle Miranda (University of Victoria no Canada)<\/strong><\/p>

T\u00edtulo: <\/strong>Modeling Modern Data Objects: Statistical Methods for Ultra-high Dimensionality and Intricate Correlation Structures.<\/p>

31\/05\/2019 \u00e0s 13:30h –\u00a0 Local: Audit\u00f3rio B 106 – CAD3<\/strong>\u00a0<\/strong><\/p>

Magda Carvalho Pires (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Current status data com censura informativa e erros de classifica\u00e7\u00e3o<\/p>

Este semin\u00e1rio \u00e9 integra a programa\u00e7\u00e3o do V Encontro Comemorativo do Dia do Estat\u00edstico. Por favor inscreva-se: (http:\/\/www.est.ufmg.br\/diadoestatistico\/inscricoes.html<\/a>)<\/p>

24\/05\/2019 \u00e0s 13:30h – CHICO Soares (Prof. Em\u00e9rito da UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Minhas Estat\u00edsticas<\/p>

17\/05\/2019 \u00e0s 13:30h – Marcos O. Prates (EST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Assessing spatial confounding in Bayesian shared component disease mapping models via SPOCK: With applications to SEER cancer data<\/p>

03\/05\/2019 \u00e0s 13:30h – Afr\u00e2nio M C Vieira (UFSCAR)<\/strong><\/p>

T\u00edtulo: <\/strong>Modelos de Resposta ao Item modificados para fontes de heterogeneidade conhecidas e desconhecidas.<\/p>

26\/04\/2019 \u00e0s 13:30h – F\u00e1bio Nogueira Demarqui (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>An unified semiparametric approach to model survival data with crossing survival curves<\/p>

12\/04\/2019 \u00e0s 13:30h – Guilherme Augusto Veloso (PG-EST)<\/strong><\/p>

T\u00edtulo: <\/strong>An\u00e1lise Bayesiana Sequencial de Dados Multivariados de Contagem<\/p>

29\/03\/2019 \u00e0s 14:00h – Frederico R. B. Cruz (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Estima\u00e7\u00e3o e Otimiza\u00e7\u00e3o em Filas e Aplica\u00e7\u00f5es<\/p>

29\/03\/2019 \u00e0s 13:00h –\u00a0<\/strong>Euloge Clovis Kenne Pagui (Universit\u00e0 di Padova, It\u00e1lia)<\/strong><\/p>

T\u00edtulo: <\/strong>Bias reducing adjusted score functions for monotone likelihood in Cox Regression<\/p>

21\/03\/2019 \u00e0s 14:00h – Nitis Mukhopadhyay (Department of Statistics –\u00a0University of Connecticut)<\/strong><\/p>

T\u00edtulo: <\/strong>On Asymptotic Normality of Standardized Stopping Times with Illustrations<\/p>

ANO DE 2018 – 2\u00ba SEMESTRE<\/strong><\/p>

07\/12\/2018 \u00e0s 13:30h – Douglas Mateus da Silva<\/strong><\/p>

T\u00edtulo: <\/strong>Estimador subsemble espacial para dados massivos em geoestat\u00edstica.<\/p>

30\/11\/2018 \u00e0s 13:30h – Juliana Vilela Bastos (Coordenadora do Programa Traumatismos Dent\u00e1rios da Faculdade de Odontologia da UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Metodologia e Estat\u00edstica na Pesquisa em Traumatismos Dent\u00e1rios<\/p>

30\/11\/2018 \u00e0s 14:30h – Profa. Jussiane Gon\u00e7alves (UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Modelagem de sobredispers\u00e3o tempo-dependente em dados de contagem longitudinal<\/p>

23\/11\/2018 \u00e0s 10:00h – Prof. Murray Pollock (Un. of Warwick)<\/strong><\/p>

T\u00edtulo: <\/strong>Modelo de regress\u00e3o de Cox com verossimilhan\u00e7a mon\u00f3tona<\/p>

23\/11\/2018 \u00e0s 13:30h – Frederico Machado Almeida<\/strong><\/p>

T\u00edtulo: <\/strong>Confusion: Developing an information-theoretic secure approach for multiple parties to pool and unify statistical data, distributions and inferences.<\/p>

23\/11\/2018 \u00e0s 14:30h – Luis Alejandro M\u00e1smela Caita<\/strong><\/p>

T\u00edtulo: <\/strong>Imputa\u00e7\u00e3o M\u00faltipla para dados ausentes de maneira n\u00e3o-aleat\u00f3ria<\/p>

09\/11\/2018 \u00e0s 13:30h – Arthur Tarso Rego<\/strong><\/p>

T\u00edtulo: <\/strong>Abordagem via Modelos de Espa\u00e7o de Estados para S\u00e9ries Temporais Financeiras<\/p>

26\/10\/2018 \u00e0s 14:30h – Guilherme Aguilar<\/strong><\/p>

T\u00edtulo: <\/strong>Bayesian linear regression models with flexible error distributions<\/p>

26\/10\/2018 \u00e0s 13:30h – Danna L. Cruz<\/strong><\/p>

T\u00edtulo: <\/strong>Spatial disease mapping using Directed Acyclic Graph Auto-Regressive (DAGAR) models<\/p>

19\/10\/2018 \u00e0s 14:30h – Profa. Thais C. O. Fonseca (DME-UFRJ)<\/strong><\/p>

T\u00edtulo: <\/strong>Reference Bayesian analysis for hierarchical models<\/p>

19\/10\/2018 \u00e0s 13:30h – Prof. Karthik Bharath (University of Nottingham, UK)<\/strong><\/p>

T\u00edtulo: <\/strong>Geometric statistical methods for imaging data<\/p>

28\/09\/2018 \u00e0s 13:30h – Larissa Sayuri Futino C. dos Santos (UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Ampliando Horizontes: Vendo o mundo com outros olhos<\/p>

14\/09\/2018 \u00e0s 13:30h – Prof. Tohid Ardeshiri (Link\u00f6ping University, Su\u00e9cia)<\/strong><\/p>

T\u00edtulo: <\/strong>Analytical Approximations for Bayesian Inference<\/p>

31\/08\/2018 \u00e0s 14:30h – Josemar Rodrigues (UFSCar)<\/strong><\/p>

T\u00edtulo: <\/strong>Bayesian superposition of pure-birth destructive cure processes for tumor latency<\/p>

3108\/2018 \u00e0s 13:30h – Reinaldo B. Arellano-Valle (Pont\u00edcia Universidad Cat\u00f3lica de Chile)<\/strong><\/p>

T\u00edtulo: <\/strong>Scale and Shape Mixtures of Multivariate Skew-Normal Distributions<\/p>

24\/08\/2018 \u00e0s 13:30h – Roger W. C. da Silva (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Dimensional Crossover in Anisotropic Percolation on Z^{d+s}<\/p>

17\/08\/2018 (sexta-feira) \u00e0s 13:30h – Ali Abolhassani (Department of Mathematical Sciences, Isfahan University of Technology, Isfahan, Iran)<\/strong><\/p>

T\u00edtulo: <\/strong>Bell Spatial Scan Statistics<\/p>

08\/08\/2018 (quarta-feira) \u00e0s 10:30h – Silvia L. P. Ferrari (USP)<\/strong><\/p>

T\u00edtulo: <\/strong>Box-Cox t random intercept model for estimating usual nutrient intake distributions<\/p>

ANO DE 2018 – 1\u00ba SEMESTRE<\/strong><\/p>

22\/06\/2018 \u00e0s 14:30h – T\u00falio Lima (Departamento de Estat\u00edstica – UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Comparison between risk measures and ruin probability for the calculation of solvency capital for a long-term guarantee.<\/p>

18\/05\/2018 \u00e0s 13:30h – Rodrigo Bernardo da Silva (Departamento de Estat\u00edstica, UFPb)<\/strong><\/p>

T\u00edtulo: <\/strong>Flexible and Robust Mixed Poisson INGARCH Models.<\/p>

11\/05\/2018 \u00e0s 13:30h – Vinicius D. Mayrink (Departamento de Estat\u00edstica, UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Estendendo o JAGS: Distribui\u00e7\u00e3o exponencial por partes e geoestat\u00edstica.<\/p>

04\/05\/2018 \u00e0s 13:30h – Caio L. N. Azevedo –\u00a0Departamento de Estat\u00edstica, IMECC, Unicamp<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Time series and multilevel modeling for longitudinal item response theory data<\/p>

27\/04\/2018 \u00e0s 13:30h – Vald\u00e9rio A. Reisen – UFES<\/strong><\/p>

T\u00edtulo: <\/strong>An overview of robust spectral estimators and its applications.<\/p>

20\/04\/2018 \u00e0s 13:30h – Pedro O. S. Vaz de Melo<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Futebol e Pol\u00edtica n\u00e3o se discutem, se analisam!<\/p>

13\/04\/2018 \u00e0s 13:30h – Carolina Silva Pena – Pr\u00f3-Reitoria de Gradua\u00e7\u00e3o – UFMG<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>A new item response theory model to adjust data allowing examinee choice<\/p>

ANO DE 2017 – 2\u00ba SEMESTRE<\/strong><\/p>

01\/12\/2017 \u00e0s 13:30h –\u00a0Milton Pifano (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Data clustering using generalized spatio-temporal dynamic factor analysis with interactions.<\/p>

24\/11\/2017 \u00e0s 13:30h – Guilherme L. de Oliveira (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Modelos Parti\u00e7\u00e3o Produto Espaciais.<\/p>

24\/11\/2017 \u00e0s 14:30h – Gabriela Oliveira (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Aspectos Probabil\u00edsticos da Distribui\u00e7\u00e3o Laplace.<\/p>

17\/11\/2017 \u00e0s 13:30h – Alexandre Gaudilli\u00e8re (Aix Marseille Universit\u00e9, CNRS)<\/strong><\/p>

T\u00edtulo: <\/strong>Intertwining Wavelets.<\/p>

17\/11\/2017 \u00e0s 14:30h – Douglas Mesquita (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Confundimento espacial em modelos de fragilidade.<\/p>

10\/11\/2017 \u00e0s 13:30h – Erick Amorim (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Agrupamentos atrav\u00e9s do processo Dirichlet e o modelo fatorial com intera\u00e7\u00f5es<\/p>

10\/11\/2017 \u00e0s 14:30h – Rafael Alves (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Markov Graphs.<\/p>

27\/10\/2017 \u00e0s 13:30h – Juliana Freitas de Mello e Silva (DEST-UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Modelagem conjunta de dados longitudinais e de sobreviv\u00eancia.<\/p>

20\/10\/2017 \u00e0s 13:30h – Fl\u00e1vio Bambirra Gon\u00e7alves (DEST-UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>A Monte Carlo toolbox to solve intractable statistical problems: from retrospective sampling to Bernoulli Factories<\/p>

29\/09\/2017 \u00e0s 13:30h – Gilvan Ramalho Guedes (Depto. De Demografia-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Mudan\u00e7as clim\u00e1ticas e economia: impactos sobre vulnerabilidade regional, oferta de trabalho e demanda por seguro<\/strong><\/p>

22\/09\/2017 \u00e0s 13:30h – Fernando Quintana (PUC-Chile)<\/strong><\/p>

T\u00edtulo: <\/strong>Covariate-Dependent Mixture Models Induced by Determinantal Point Processes and Some Applications<\/p>

15\/09\/2017 \u00e0s 13:30h – Grupo Stats4Good (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Estat\u00edstica para o Bem<\/p>

01\/09\/2017 \u00e0s 13:30h – Thais Paiva (DEST-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>Imputation of multivariate continuous data with nonignorable missingness<\/p>

25\/08\/2017 \u00e0s 13:30h –\u00a0Bernardo Nunes Borges de Lima (MAT-UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>A m\u00e1gica sequ\u00eancia de de Bruijn<\/p>

18\/08\/2017 \u00e0s 11:10h – Sokol Ndreca (DEST)<\/strong><\/p>

T\u00edtulo: <\/strong>Asymptotics for the queueing system with exponentially delayed arrivals<\/p>

16\/08\/2017 \u00e0s 11:30h (excepcionalmente) –\u00a0Iddo Ben-Ari (University of Connecticut – USA)<\/strong><\/p>

T\u00edtulo: <\/strong>Cut-off for a random walk with catastrophes<\/p>

ANO DE 2017 – 1\u00ba SEMESTRE<\/strong><\/p>

11\/08\/2017 \u00e0s 14:30h – Ying Sun (King Abdullah University of Science and Technology (KAUST),Saudi Arabia)<\/strong><\/p>

T\u00edtulo: <\/strong>Visualization and Assessment of Spatio-temporal Covariance Properties<\/p>

11\/08\/2017 \u00e0s 13:30h –\u00a0Marc G. Genton (King Abdullah University of Science and Technology (KAUST), Saudi Arabia)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Directional Outlyingness for Multivariate Functional Data<\/p>

07\/07\/2017 \u00e0s 13:30h – B\u00e1rbara da Costa Campos Dias<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Exact Bayesian inference in spatio-temporal Cox processes driven by multivariate Gaussian processes<\/p>

30\/06\/2017 \u00e0s 13:30h – Uriel Moreira Silva<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Particle-based Inferente in Hidden Markov Models<\/p>

23\/06\/2017 \u00e0s 13:30h – Prof.\u00a0Alexandre B. Simas (MAT-UFPb)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Principal Components Analysis for Semimartingales and Stochastic PDE<\/p>

09\/06\/2017 \u00e0s 13:30h – Prof. Adrian P. H. Luna (DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Misturas de Distribui\u00e7\u00f5es de Gibbs<\/p>

02\/06\/2017 \u00e0s 13:30h – Prof. Bernardo Lanza Queiroz\u00a0(CEDEPLAR\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>National and subnational experience with estimating the extent and trend in completeness of registration of deaths in Brazil and other\u00a0developing countries<\/p>

19\/05\/2017 \u00e0s 13:15h (**Excepcionalmente**) – Prof. Fredy Castellares (DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0\u00a0<\/strong>Processo M\u00faltiplo de Poisson e a Distribui\u00e7\u00e3o de Bell<\/p>

28\/04\/2017 \u00e0s 13:15h (**Excepcionalmente**) – Prof. Bernardo Nunes Borges de Lima (MAT\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0 <\/strong>A m\u00e1gica sequ\u00eancia de Bruijn<\/p>

07\/04\/2017 \u00e0s 13:30h – Prof. Marcos Oliveira Prates (DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Um passeio por aplica\u00e7\u00f5es e problemas em diferentes \u00e1reas da Estat\u00edstica nas quais tenho dedicado o meu tempo.<\/p>

31\/03\/2017 \u00e0s 13:30h – Profa. Denise Duarte (DEST\/UFMG)<\/strong><\/p>

T\u00edtulo:\u00a0<\/strong>Infer\u00eancia para Cadeias de Markov de Alcance Vari\u00e1vel Contaminadas Estocasticamente<\/p>

24\/03\/2017 \u00e0s 13:30h – Prof. Renato Martins Assun\u00e7\u00e3o (DCC\/UFMG)<\/strong><\/p>

T\u00edtulo: <\/strong>De Fisher at\u00e9 o “Big Data”: continuidades e descontinuidades<\/p>\t\t\t\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t

\n\t\t\t\t
\n\t\t\t\t\t
\"\"<\/figure>

Vin\u00edcius Diniz Mayrink <\/h3>

Departamento de Estat\u00edstica - UFMG<\/p><\/div><\/div>\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/div>\n\t\t\t\t\t<\/div>\n\t\t<\/section>\n\t\t\t\t<\/div>\n\t\t","protected":false},"excerpt":{"rendered":"

Semin\u00e1rios do DEST ANO DE 2026 – 1\u00ba SEMESTRE 17\/04\/2026 \u00e0s 13:30hs \u2013 Local: sala 2076 – ICEx Luiz Henrique Duczmal (DEST-UFMG) T\u00edtulo: Particle Manifold Metropolis-adjusted Langevin Algorithms Resumo: We propose a tree-spatial scan statistic that combines Kulldorff\u2019s circular scan method\u00a0for detecting spatial clusters and the tree-based scan statistic algorithm for data mining. We feed […]<\/p>\n","protected":false},"author":35,"featured_media":0,"parent":0,"menu_order":0,"comment_status":"closed","ping_status":"closed","template":"","meta":{"site-sidebar-layout":"no-sidebar","site-content-layout":"page-builder","ast-site-content-layout":"full-width-container","site-content-style":"unboxed","site-sidebar-style":"unboxed","ast-global-header-display":"","ast-banner-title-visibility":"","ast-main-header-display":"","ast-hfb-above-header-display":"","ast-hfb-below-header-display":"","ast-hfb-mobile-header-display":"","site-post-title":"disabled","ast-breadcrumbs-content":"","ast-featured-img":"disabled","footer-sml-layout":"","ast-disable-related-posts":"","theme-transparent-header-meta":"default","adv-header-id-meta":"","stick-header-meta":"","header-above-stick-meta":"","header-main-stick-meta":"","header-below-stick-meta":"","astra-migrate-meta-layouts":"set","ast-page-background-enabled":"default","ast-page-background-meta":{"desktop":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"ast-content-background-meta":{"desktop":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"tablet":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""},"mobile":{"background-color":"var(--ast-global-color-5)","background-image":"","background-repeat":"repeat","background-position":"center center","background-size":"auto","background-attachment":"scroll","background-type":"","background-media":"","overlay-type":"","overlay-color":"","overlay-opacity":"","overlay-gradient":""}},"footnotes":""},"class_list":["post-13819","page","type-page","status-publish","hentry"],"_links":{"self":[{"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/pages\/13819","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/pages"}],"about":[{"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/types\/page"}],"author":[{"embeddable":true,"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/users\/35"}],"replies":[{"embeddable":true,"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/comments?post=13819"}],"version-history":[{"count":280,"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/pages\/13819\/revisions"}],"predecessor-version":[{"id":21933,"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/pages\/13819\/revisions\/21933"}],"wp:attachment":[{"href":"https:\/\/www.est.ufmg.br\/portal\/wp-json\/wp\/v2\/media?parent=13819"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}