Estimação de maxima verossimilhança para modelos de sobrevivencia com efeitos aleatorios

AUTOR(ES)
DATA DE PUBLICAÇÃO

1999

RESUMO

The goat of this work is to verify the consistency and asymptotic normality of one estimator suggested in the literature regarding the variance of the random effect (frailty), assuming that the distribution of the frailty follows one of the three models: gamma, lognormal and normal. It is of great interest to verify consistency and asymptotic normality of an estimator since these are the properties we would need if we were interested in testing whether the population is homogeneous or not. Through the simulation results we verify that the estimator of the variance of the random effect is very sensitive to the choice of the distribution. If the frailty is gamma-distributed, the estimator is consistent and asymptotically normal, however for the lognormal distribution it underestimate the variance and under the normal model it overestimate the variance. Moreover, we verify that the estimation of the cumulative risk function is not good for all the models

ASSUNTO(S)

bioestatistica

Documentos Relacionados