A0451
Title: A weighted partial likelihood approach for zero-truncated models
Authors: Jakub Stoklosa - University of New South Wales (Australia) [presenting]
Wen-Han Hwang - National Tsing Hua University (Taiwan)
Abstract: Motivated by the Rao-Blackwell theorem, we develop a weighted partial likelihood approach to estimate model parameters for the zero-truncated binomial distribution. The resulting estimating function is equivalent to a weighted score function for a standard binomial model, hence allowing for straightforward implementation for estimating model parameters. We evaluate the efficiency for this new approach and show that it performs almost as well as the maximum zero-truncated likelihood method. In addition, the weighted partial likelihood approach can also be extended to zero-truncated Poisson models. An application to estimating population sizes using capture-recapture models is also addressed. This novel approach is then implemented with a corrected score method to accommodate models with measurement error, which has yet to be developed for zero-truncated regression measurement error models. We examine the performance of the proposed methods through simulation studies and real data.
