Variable Selection of Interval-Censored Failure Time Data
Joint Biostatistics Seminar
(Tony) Jianguo Sun, Ph.D.
Department of Statistics, University of Missouri
Variable selection is a commonly asked question in statistical analysis and has been extensively discussed under many contexts. In particular, many authors have investigated the problem under the survival analysis context. However, most of the existing methods for failure time data only deal with right-censored data and in this talk, we will discuss the problem for regression analysis of interval-censored data, a more general type of failure time data, under the proportional hazards model. For covariate selection and estimation of covariate effects, we will present some penalized estimation procedures and discuss their applications in practice.