Variable Selection for Survival Data Analysis

• Activity Number: 175
• Type: Invited
• Date/Time: Tuesday, August 10, 2004 : 8:30 AM to 10:20 AM
• Sponsor: Section on Statistical Computing
• Author(s): Runze Li*+ and Jianwen Cai and Jianqing Fan and Haibo Zhou
• Companies: Pennsylvania State University and University of North Carolina, Chapel Hill and Princeton University and University of North Carolina, Chapel Hill
• Keywords: penalized likelihood ; SCAD

Abstract

In most model selection problems, the number of parameters may be large, and grows as the sample size increases. I propose a penalized-likelihood method for variable selection with survival data analysis with growing number of regression coefficients. Under certain regularity conditions, we show the consistency and asymptotic normality of the penalized-likelihood estimators. We further demonstrate that, for certain penalty functions with proper choices of regularization parameters, the resulting estimate possesses an oracle property, namely, the resulting estimate can correctly identify the true model as if the true model (the subset of variables with nonvanishing coefficients) were known in advance. Using a simple approximation of the penalty function, the proposed method can be easily carried out with the Newton-Raphson algorithm. We conduct extensive Monte Carlo simulation studies to assess the finite sample performance of the proposed procedures. We illustrate the proposed method by analyzing a real dataset.