|
Abstract:
|
Interval-censored data naturally arise many epidemiologic, social- behavioral, and medical studies, in which subjects are examined multiple times and the failure times of interest are not observed exactly, but fall within some intervals. Two different frailty probit models are proposed for the regression analysis of multivariate interval-censored data, and both models allow explicit form of the pairwise statistical association among the failure times. Two efficient Bayesian estimation approaches are proposed under the two models and allow joint estimation of regression parameters and other secondary parameters. The two proposed methods are evaluated by extensive simulation studies and illustrated by two real-life applications.
|
