Sufficient Dimension Reduction for Censored Regressions
Article first published online: 28 SEP 2010
© 2010, The International Biometric Society
Volume 67, Issue 2, pages 513–523, June 2011
How to Cite
Lu, W. and Li, L. (2011), Sufficient Dimension Reduction for Censored Regressions. Biometrics, 67: 513–523. doi: 10.1111/j.1541-0420.2010.01490.x
- Issue published online: 20 JUN 2011
- Article first published online: 28 SEP 2010
- Received July 2009. Revised May 2010. Accepted July 2010.
- Censored data;
- Central subspace;
- Inverse censoring probability weighted estimation;
- Sliced inverse regression;
- Sufficient dimension reduction
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.