American Statistical Association New York City Metropolitan Area Chapter Mailman School of Public Health Columbia University Department of Biostatistics Colloquium

In this talk, we propose the partly proportional single-index hazards model for right censored survival data. This model is an extension of the Cox proportional hazards model and allows flexible semiparametric modeling of covariate effects in a parsimonious way via a single-index. The commonly used profile-kernel method based on the local likelihood is considered for parameter estimation. We show that this popular approach leads to biased estimation. A bias correction method is then proposed and the corrected profile local likelihood estimators are shown to be consistent, asymptotically normal and semiparametrically efficient. We evaluate the finite-sample properties of our estimators through simulation studies and illustrate the proposed model and method with an application to a dataset from the Multicenter AIDS Cohort Study (MACS).

Donglin Zeng is currently an associate professor in the Department of Biostatistics at the University of North Carolina at Chapel Hill. He obtained his Ph.D. from the Department of Statistics at the University of Michigan in 2001. He won the Noether Young Scholar award in 2008 and was elected an IMS fellow in 2010. His primary research interests are semiparametric models and high-dimensional inference in survival analysis, longitudinal data analysis, and genetic epidemiology. His other collaborations involve cancer studies, breast imaging, and AIDS studies.