Sufficient Dimension Reduction for Censored Regressions

Authors

  • Wenbin Lu,

    Corresponding author
    1. Department of Statistics, North Carolina State University, Raleigh, North Carolina 27695, U.S.A.
      email: lu@stat.ncsu.edu
    Search for more papers by this author
  • Lexin Li

    Corresponding author
    1. Department of Statistics, North Carolina State University, Raleigh, North Carolina 27695, U.S.A.
      email: li@stat.ncsu.edu
    Search for more papers by this author

email:lu@stat.ncsu.edu

email:li@stat.ncsu.edu

Abstract

Summary Methodology of sufficient dimension reduction (SDR) has offered an effective means to facilitate regression analysis of high-dimensional data. When the response is censored, however, most existing SDR estimators cannot be applied, or require some restrictive conditions. In this article, we propose a new class of inverse censoring probability weighted SDR estimators for censored regressions. Moreover, regularization is introduced to achieve simultaneous variable selection and dimension reduction. Asymptotic properties and empirical performance of the proposed methods are examined.

Ancillary