Advanced search
Start date

Estimation in mixed-effects models with censored response using scale mixtures of normal distributions

Grant number: 15/05385-3
Support type:Scholarships abroad - Research Internship - Doctorate
Effective date (Start): July 15, 2015
Effective date (End): January 14, 2016
Field of knowledge:Physical Sciences and Mathematics - Probability and Statistics - Statistics
Principal Investigator:Víctor Hugo Lachos Dávila
Grantee:Larissa Avila Matos
Supervisor abroad: Ming-Hui Chen
Home Institution: Instituto de Matemática, Estatística e Computação Científica (IMECC). Universidade Estadual de Campinas (UNICAMP). Campinas , SP, Brazil
Local de pesquisa : University of Connecticut (UCONN), United States  


This project will present a new approach for estimation in mixed-effects models with censored re- sponse using scale mixtures of normal distributions. In the framework of linear and non-linear mixed- effects censored (LMEC/NLMEC) models, the random effects and residual errors are routinely assumed to have a normal distribution, mainly for mathematical convenience. However, this method has been criticized in the literature due to its sensitivity to deviations from the normality assumption. In this project, we will assume that the random effects and errors jointly follow a multivariate scale mixture of normal distributions. The Student-t, slash and contaminated normal, among others distributions, are contained in this class and provide interesting alternatives to the Gaussian model, which produce robust estimates against outlying observations. We will propose an exact estimation procedure to obtain the maximum likelihood estimates of the model parameters, using a Stochastic Approximation of the traditional EM algorithm proposed by Dempster et al. (1977), called SAEM (Delyon et al., 1999). The main goal is to extend the work of Matos et al. (2013), Meza et al. (2012) and Samson et al. (2006). (AU)

Scientific publications
(References retrieved automatically from Web of Science and SciELO through information on FAPESP grants and their corresponding numbers as mentioned in the publications by the authors)
MATOS, LARISSA A.; LACHOS, VICTOR H.; LIN, TSUNG-I; CASTRO, LUIS M. Heavy-tailed longitudinal regression models for censored data: a robust parametric approach. TEST, v. 28, n. 3, p. 844-878, SEP 2019. Web of Science Citations: 0.
MATOS, LARISSA A.; CASTRO, LUIS M.; CABRAL, CELSO R. B.; LACHOS, VICTOR H. Multivariate measurement error models based on Student-t distribution under censored responses. STATISTICS, v. 52, n. 6, p. 1395-1416, 2018. Web of Science Citations: 0.

Please report errors in scientific publications list by writing to: