Abstract:
|
Recurrent event data are important in many fields, such as biomedical science, econometrics, reliability, and demography. In many applications, recurrent events of study subject serve as important measurements for evaluating disease progression, health deterioration, or insurance risk. In this talk, I will present a joint model of the recurrent event process and the failure time where a common latent variable is used to induce the association between the intensity of the recurrent event process and the hazard of the failure time. The distributions of the censoring variable and the latent variable are treated as nuisances. The censoring mechanism is allowed to be correlated with the recurrent event process and the failure time. A `borrow-strength' method is adopted by first estimating the subject-specific latent variable value from the recurrent event data, and next using the estimated value of the latent variable in the failure time models. Properties of the regression parameters and baseline cumulative hazard functions are explored. A Denmark schizophrenia cohort data is used to illustrate the proposed model.
|