Efficient score estimation and adaptive M-estimators in censored and truncated regression models

An adaptive hi-estimator of a regression parameter based on censored and tr uncated data is developed by using B-splines to estimate the efficient scor e function and a relatively simple cross validation method to determine the number of knots. An iterative algorithm to compute the estimator is also p rovided. The adaptive estimator is asymptotically efficient, and simulation studies of the finite-sample performance of the adaptive estimator shows t hat it is superior to other M- estimators for regression analysis of censor ed and truncated data in the literature. An asymptotic theory of cross vali dation in the presence of censoring and truncation is also developed in thi s connection.