论文部分内容阅读
HERE-One difficulty in analyzing length-biased data is the informative right centring due to the sampling mechanism.We evaluate covariate effects on the failure times of the target population under a semiparametric accelerated failure time (AFT) model, given the observed length-biased data.The AFT model structure changes under length-biased sampling when there is right censoring, and the techniques for conventional survival analysis are not applicable.We develop two estimating equation approaches to estimate the covariate effects on the unbiased failure times.