Speaker:
Abstract: Akaike Information Criterion (AIC), which is based on maximum likelihood estimation but cannot be applied directly to the situations when likelihood functions are not available, has been modified for variable selection in longitudinal data with generalized estimating equations via working independence model in the literature. In this talk, I will present a variant version, the difference between the quasi-likelihood functions of a candidate model and of a narrow model plus a penalty term. Such a difference avoids calculating complex integration from quasi-likelihood, but inherits theoretical asymptotic properties from AIC. We also propose a focused information criterion for variable selection on the basis of quasi-score function. Simulation studies provide evidence on the superiority of the proposed procedures. The procedures are further applied to a real data example. Speaker Bio: Dr. Liang received his Ph.D. in Mathematical Statistics from the Institute of Systems Science, Chinese Academy of Sciences in 1992, and Ph.D. in statistics from Texas A&M University in 2001. He was Assistant Professor (2002-2005) at St. Jude Children's Research Hospital and Associate Professor (2005-2009) and Professor (2009-2013) at the University of Rochester Medical Center. Dr. Liang has worked on semi-parametric regression, mixed-effects models for longitudinal data, missing data, measurement error models, variable selection, and HIV dynamic models. He has been awarded two RO1, one T32, and five NSF grants. He is a Fellow of ASA, IMS, Royal Statistical Society, and a member of the International Statistical Institute. Dr. Liang has served as an Associate Editor for Biostatistics, Electronic Journal of Statistics, and JASA.
Zoom Link: https://bit.ly/3sUHYmm Passcode: 801063