Quasi-Likelihood-Based Focused Information Criterion and Frequentist Model Averaging for Longitudinal Data

• Activity Number: 428
• Type: Contributed
• Date/Time: Tuesday, August 6, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Biopharmaceutical Section
• Author(s): Hui Yang*+ and Guohua Zou and Hua Liang
• Companies: and Chinese Academy of Sciences and University of Rochester
• Keywords: Focus parameter ; Local misspecification ; Quasi-likelihood ; Quasi-score function ; Smoothing QFIC ; Model averaging

Abstract

In this paper, we study model selection and model averaging for longitudinal data when their distribution is unnecessarily specified. We develop a focused information criterion (QFIC) and a frequentist model average (QFMA) estimator on the basis of quasi-score function. We further examine the impacts of various weights on model averaging estimates, and give a suggestion for what weight is preferable from numerical prospects. Simulation studies provide evidence on the superiority of the proposed procedures. The procedure is further applied to a real data example.