Transition models for binary data:
tests to compare treatments

Idemauro Antonio Rodrigues de LARA[1]

Clarice Garcia Borges DEMÉTRIO1

Dalton Francisco de ANDRADE[2]

João Maurício Araújo MOTA[3]

§    ABSTRACT: This work focus on generalized linear transition models suitable for analyzing longitudinal data with binary response. Such models are based on stochastic processes and we aim to model the probabilities of change or transitions of individual response categories in time. The maximum likelihood approach is used in order to fit the models and estimate the probabilities. Furthermore, we propose asymptotic tests to compare treatments based on odds ratio and on the differences of transition probabilities. The methods are illustrated with respiratory disease data. For these data, the process is stationary of order two and the suggested test points to a significant statistical difference in favour of the active treatment.

§    KEYWORDS: Longitudinal data; generalized linear model; stochastic processes; transition probabilities; maximum likelihood.

 



[1] Departamento de Ciências Exatas, Escola Superior de Agricultura "Luiz de Queiroz" – Universidade de São Paulo – ESALQ/USP, CEP: 13418-900, Piracicaba, São Paulo, Brasil. E-mail: idemauro@esalq.usp.br / clarice@esalq.usp.br

[2] Departamento de Informática, Universidade Federal de Santa Catarina – UFSC, CEP: 88040-970, Florianópolis, SC, Brasil. E-mail: dandrade@inf.ufsc.br

[3] Departamento de  Estatística, Universidade Federal do CEARÁ – UFC, CEP: 60020-181, Fortaleza, CE, Brasil. E-mail: oiciruam@ufc.br