Nonlinear mixed-effects models with non-normally distributed random components

• Grant number: 12/22696-4
• Support Opportunities: Scholarships abroad - Research
• Start date: May 20, 2013
• End date: March 19, 2014
• Field of knowledge: Physical Sciences and Mathematics - Probability and Statistics - Statistics
• Principal Investigator: Cibele Maria Russo Novelli
• Grantee: Cibele Maria Russo Novelli
• Host Investigator: Emmanuel Lesaffre
• Host Institution: Instituto de Ciências Matemáticas e de Computação (ICMC). Universidade de São Paulo (USP). São Carlos , SP, Brazil
• Institution abroad: Erasmus University Rotterdam (EUR), Netherlands

Abstract

The importance of mixed-effects models is unquestionable for problems with correlated data, as longitudinal data or repeated measurements. Linear mixed-effects models have been extensively studied with a rich literature as a result. Also, generalized linear mixed models have been investigated quite intensively. However, non-linear mixed models, essential in areas as pharmacokinetics, growth curves or industrial data, have been less explored. These models are also notoriously more complex to apply in practice. In this application we are interested in the development of nonlinear mixed effects models with random components that have a non-normal distribution. Several applications with real data sets will be considered to guide and illustrate our developments. At the Erasmus Medical Center (Rotterdam), a great variety of data bases are in development or are available which can be useful in this context. One example is the PREDICT data base in which multivariate echographic data are collected of the foetus and the new born child of about 1500 pregnant women. This is an example of a multivariate growth curve model where the purpose is to establish reference ranges of the measured parameters and hence also to detect risk factors for malformations. As a conclusion, we wish to state that we intend to develop useful models and analyses for statistical and application areas. (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)

BARROS GOMES, JOSE CLELTO; AOKI, REIKO; LACHOS, VICTOR HUGO; PAULA, GILBERTO ALVARENGA; RUSSO, CIBELE MARIA. Fast inference for robust nonlinear mixed-effects models. Journal of Applied Statistics, v. N/A, p. 24-pg., 2022-02-11. (12/22696-4)

WILLEMSEN, STEN P.; RUSSO, CIBELE M.; LESAFFRE, EMMANUEL; LEAO, DORIVAL. Flexible multivariate nonlinear models for bioequivalence problems. STATISTICAL MODELLING, v. 17, n. 6, p. 449-467, DEC 2017. (12/22696-4)