Regression analysis of doubly truncated data based on pseudo-observations

摘要: Doubly truncated data arise when an individual is potentially observed only if its failure-time lies within a certain interval, unique to that individual. In this paper, we consider the pseudo-observations approach for estimating regression coefficients subject double truncation. The generated from nonparametric maximum likelihood estimates (NPMLE) of survival function are used as response variables in generalized equation estimate parameters model. We look at two estimators probabilities based on different ways defining pseudo-observations, namely, simple (SPO) and stopped (STPO). establish asymptotic properties under some conditions. Simulations results show proportion failed estimation STPO smaller than SPO. estimator performs adequately finite samples while SPO can be very unstable sample size not large enough.