Least absolute deviations estimation for the accelerated failure time model

Jian Huang, Shuangge Ma, Huiliang Xie

Research output: Journal article publicationJournal articleAcademic researchpeer-review

27 Citations (Scopus)

Abstract

The accelerated failure time (AFT) model assumes a linear relationship between event time and covariates. We propose a robust weighted least-absolutedeviations (LAD) method for estimation in the AFT model with right-censored data. This method uses the Kaplan-Meier weights in the LAD objective function to account for censoring. We show that the proposed estimator is root-n consistent and asymptotically normal under appropriate assumptions. It can also be easily computed using existing software, which makes it especially useful for data with medium to high dimensional covariates. The proposed method is evaluated using simulations and demonstrated on two clinical data sets.
Original languageEnglish
Pages (from-to)1533-1548
Number of pages16
JournalStatistica Sinica
Volume17
Issue number4
Publication statusPublished - 1 Oct 2007
Externally publishedYes

Keywords

  • Asymptotic normality
  • Kaplan-Meier weights
  • Least absolute deviations
  • Right censored data
  • Robust regression

ASJC Scopus subject areas

  • Statistics and Probability
  • Statistics, Probability and Uncertainty

Cite this