M-estimation of the accelerated failure time model under a convex discrepancy function |
| |
Authors: | Weihua Zhou |
| |
Institution: | Department of Mathematics and Statistics, University of North Carolina at Charlotte, Charlotte, NC 28223, USA |
| |
Abstract: | Based on the Kaplan–Meier weight functions, we introduce a class of M-estimators of regression parameters for the accelerated failure time (AFT) model with right censored data. The proposed M-estimator is root-n consistent and asymptotically normal under appropriate assumptions. For robustness analysis, we also derive the corresponding influence functions. Appropriate criteria are developed for tests of hypotheses concerning regression parameters. The results are applied to several particular cases. We evaluate the finite-sample performance of the proposed methods through extensive simulation studies. |
| |
Keywords: | M-estimator Accelerated failure time model Asymptotic normality Kaplan&ndash Meier weights Right censored |
本文献已被 ScienceDirect 等数据库收录! |