Abstract: This article discusses regression analysis of length-biased and rightcensored failure time data arising from the accelerated failure time model. A key
feature of such data is the informative censoring induced by the length-biased sampling, and several methods have been proposed in the literature for their analysis. However, these may be less efficient or apply only to limited situations. We propose a kernel-smoothed composite likelihood method for estimation of covariate effects. The proposed estimators are proved to be consistent and asymptotically normal. Simulation studies conducted to assess the finite sample performance of the method suggest that it works well for practical situations. An illustrative example is provided.

Key words and phrases: AFT model, composite likelihood, length-biased, rightcensored.