Academic Positions

  • Present 2012

    Adjunct Professor

    UFMG, Department of Statistics

  • 2011 2009

    Assistent Professor

    UFOP, Department of Mathematics

Education & Training

  • Ph.D. 2011

    Ph.D. in Statistics

    UFMG

  • M.B.A.2007

    Master in Statistics

    UFMG

  • B.A.2005

    Bachelor in Statistics

    UFMG

Honors, Awards and Grants

  • 2011
    UFMG Thesis Prize.
    Best thesis of the UFMG Statistics graduate program in the year 2011.
  • 2014-2016
    Universal Fapemig
    Universal Fapemig Grant.

Laboratory Personel

Denise Duarte

Researcher

+ Follow

Sokol Ndreca

Researcher

+ Follow

Roger Silva

Researcher

+ Follow

Pedro Araújo

Master Student

+ Follow

Probability Lab!

The Probability Lab started its activities in 2017. The lab is located at room 3029 in the Instituto de Ciências Exatas, UFMG.

Research Projects

  • image

    Topics in Long Range Percolation

    We study percolation of words on the graph Z^d with extra long range bonds.

    Researchers involved are:

    Roger W. C. Silva, Statistics Department, UFMG.

    Rémy P. Sanchis, Mathematics Department, UFMG.

    Bernardo N. B. de Lima, Mathematics Department, UFMG.

  • image

    Local Bak-Sneppen

    We study a local version of the Bak-Snepen model on any finite connected graph.

    Researchers involved are:

    Roger W. C. Silva, Statistics Department, UFMG.

    Iddo Ben-Ari, Mathematics Department, University of Connecticut .

Filter by type:

Sort by year:

Dimensional crossover in anisotropic percolation on Z^{d+s}

Rémy Sanchis, Roger W. C. Silva
Journal Paper Submitted

Abstract

We consider bond percolation on Z^d x Z^s where edges of Z^d are open with probability p < p_c(Z^d) and edges of Z^s are open with probability q, independently of all others. We obtain bounds for the critical curve in (p,q), with p close to the critical threshold p_c(Z^d). The results are related to the so-called dimensional crossover from Z^d to Z^{d+s}.

On a local version of the Bak-Sneppen model

Iddo Ben-Ari, Roger W. C. Silva
Journal Paper Submitted

Abstract

A major difficulty in studying the Bak-Sneppen model is in effectively comparing it with well-understood models. This stems from the use of two geometries: complete graph geometry to locate the global fitness minimizer, and graph geometry to replace the species in the neighborhood of the minimizer. We present a variant in which only the graph geometry is used. This allows to obtain the stationary distribution through random walk dynamics. We also show that for constant-degree graphs, the stationary fitness distribution converges to an IID law as the number of vertices tends to infinity. The notion of an a-avalanche is introduced and asymptotic results about its duration are obtained.

Shannon entropy and Kullback Leibler divergence in multivariate log fundamental skew normal and related distributions

Marina M. Queiroz, Roger W. C. Silva, Rosangela H. Loschi
Journal Paper The Canadian Journal of Statistics, Volume 44, Issue 2, June 2016, Pages 219-237

Abstract

This article mainly focuses on studying the Shannon entropy and Kullback–Leibler divergence of the multivariate log-canonical fundamental skew-normal (LCFUSN) and canonical fundamental skew-normal (CFUSN) families of distributions, extending previous works. We relate our results with entropies of other well-known distributions. As a by-product we also obtain the mutual information for distributions in these families. Shannon entropy is used to compare models fitted to analyze monthly USA precipitation data. Kullback–Leibler divergence is used to cluster regions in the Atlantic ocean according to their air humidity level.

Multivariate log-skewed distributions with normal kernel and their applications

Marina M. Queiroz, Rosangela H. Loschi, Roger W. C. Silva
Journal Paper Statistics (Berlin), Volume 50, Issue 1, February 2016, Pages 157-175

Abstract

We introduce two classes of multivariate log-skewed distributions with normal kernel: the log canonical fundamental skew-normal (log-CFUSN) and the log unified skew-normal. We also discuss some properties of the log-CFUSN family of distributions. These new classes of log-skewed distributions include the log-normal and multivariate log-skew normal families as particular cases. We discuss some issues related to Bayesian inference in the log-CFUSN family of distributions, mainly we focus on how to model the prior uncertainty about the skewing parameter. Based on the stochastic representation of the log-CFUSN family, we propose a data augmentation strategy for sampling from the posterior distributions. This proposed family is used to analyse the US national monthly precipitation data. We conclude that a high-dimensional skewing function lead to a better model fit.

Modelando a taxa de neoplasia pulmonar no Brasil via modelos lineares generalizados

Thiago R. dos Santos, Roger W. C. Silva
Journal Paper Revista da Estatística da Universidade Federal de Ouro Preto, Volume 3, Issue 1, April 2014, Pages 1-6

Abstract

O câncer de pulmão é uma das principais causas de morte na atualidade em todo mundo, tendo como um de seus principais fatores de risco o tabagismo, eassim, constitui um dos mais importantes problemas de saude pública no Brasil. Neste trabalho, utilizando dados correspondentes a taxa de incidência da doença nas cinco regiões brasileiras, discriminadas por sexo e ano, ajustamos um Modelo Linear Generalizado para modelar a variável resposta Taxa de Incidência de Neoplasia Pulmonar através das covariaveis mencionadas, a saber, Região, Sexo e Ano. Mostramos que o modelo é adequado e que todas as covariáveis são significativas aonível de 5% de significância com exceção da variável Ano. As regiões sul e sudeste têm as maiores taxas e os homens apresentam umapropensão maior que as mulheres.

Critical point and percolation probability in a long range site percolation model on Z^d

Bernardo N. B. de Lima, Rémy Sanchis, Roger W. C. Silva
Journal Paper Stochastic Process and Their Applications, Volume 121, Issue 9, September 2011, Pages 2043-2048

Abstract

Consider an independent site percolation model with parameter p ϵ (0,1) on Z^d, d≥2, where there are only nearest neighbor bonds and long range bonds of length k parallel to each coordinate axis. We show that the percolation threshold of such a model converges to p_c(Z^{2d}) when k goes to infinity, the percolation threshold for ordinary (nearest neighbor) percolation on Z^{2d}. We also generalize this result for models whose long range bonds have several lengths.

Percolation of words on Z^d with long-range connections

Bernardo N. B. de Lima, Rémy Sanchis, Roger W. C. Silva
Journal Paper Journal of Applied Probability, Volume 48, Issue 4, December 2011, Pages 1152-1162

Abstract

Consider an independent site percolation model on Z^d with parameter p ϵ (0,1), where all long-range connections in the axis directions are allowed. In this work we show that, given any parameter p, there exists an integer K(p) such that all binary sequences (words) ξ∈ {0,1}^N can be seen simultaneously, almost surely, even if all connections with length larger than K(p) are suppressed. We also show some results concerning how K (p) should scale with p as p goes to 0. Related results are also obtained for the question of whether or not almost all words are seen.

Currrent Teaching

  • Present

    EST851 - Probabilidade

    Graduate course.

  • Present

    EST031 - Estatística e Probabilidade

    Undergraduate course.

Teaching History

  • 2016/2

    EST055 - Inferência

    Undergraduate course

  • 2016/2

    EST072 - Estatística Econômica I

    Undergraduate course.

  • 2016/1

    EST045 - Probabilidade e Processos Estocásticos

    Undergradute course.

  • 2016/1

    EST028 - Probabilidade II

    Undergraduate course.

  • 2015/2

    EST856 - Probabilidade Avançada

    Graduate course.

  • 2015/2

    EST032 - Probabilidade

    Undergraduate course.

    2015/1

    EST031 - Estatística e Probabilidade

    Undergraduate course.

    2015/1

    EST865 - Processos Estocásticos

    Graduate course.

    2014/2

    EST856 - Probabilidade Avançada

    Graduate course.

    2014/2

    EST045 - Probabilidade e Processos Estocásticos

    Undergraduate course.

    2014/1

    EST0184 - Iniciação à Estatística

    Undergraduate course.

    2014/1

    EST084 - Estatística Aplicada à Psicologia

    Undergraduate course.

    2013/2

    EST865 - Processos Estocásticos

    Graduate course.

    2013/2

    EST045 - Probabilidade e Processos Estocásticos

    Undergraduate course.

    2013/1

    EST865 - Tópicos em Cadeias de Markov

    Graduate course.

    2013/1

    EST851 - Probabilidade

    Graduate course.

    2013/1

    EST084 - Estatística Aplicada à Psicologia

    Undergraduate course.

    2012/2

    EST032 - Probabilidade

    Undergraduate course.

    2012/1

    EST851 - Probabilidade

    Graduate course.

    2012/1

    EST865 - Estatística Econômica 2

    Undergraduate course.