Statistics and Its Interface

Volume 12 (2019)

Number 1

Pseudo likelihood estimation for the additive hazards model with data subject to left-truncation and right-censoring

Pages: 135 – 148



Li-Pang Chen (Department of Statistics and Actuarial Science, University of Waterloo, Ontario, Canada)


Analysis of left-truncated and right-censored (LTRC) survival data has received extensive interest. Many inference methods have been developed for the various survival models, including the Cox proportional hazards model and the transformation model. The additive hazards model is also concerned in survival analysis, and several methods have also been developed without left-truncation. However, little work has been available in the literature for the additive hazards model with left-truncation and right-censoring. In this paper, we explore this important problem under the additive hazards model. We develop the pseudo-likelihood inference for the estimation of the survival model parameters, which yields a more efficient estimator. Besides, we assess the performance of our proposed methods using simulation studies. Through the conducted simulations, the proposed estimator is further found to outperform the existing competitors in the literature.


kernel estimator, left-truncation, misspecification, prevalent sampling, pseudo likelihood

Received 25 September 2017

Published 26 October 2018