On the Asymptotic Behavior of the Pseudolikelihood Ratio Test Statistic with Boundary Problems in Semiparametric Models

• Activity Number: 240
• Type: Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biometrics Section
• Author(s): Yong Chen*+ and Jing Ning and Kung-Yee Liang
• Companies: The University of Texas School of Public Health and The University of Texas MD Anderson Cancer Center and National Yang Ming University
• Keywords: Likelihood ratio test ; multivariate survival model ; pseudolikelihood ; semiparametric models

Abstract

In this talk, we consider the asymptotic distribution of the likelihood ratio statistic $T$ for hypothesis testing under semiparameter models based on the pseudolikelihood (Gong and Samaniego, 1981). The pseudolikelihood is contructed as $L(\theta, \hat \phi)$, where $\hat \phi$ is a consistent estimator of the infinite dimensional nuisance parameter $\phi$. We show that the asymptotic distribution of $T$ under $H_0$ is a weighted sum of independent $\chi_1^2$ variables. Examples in multivariate survival analysis are given.