On a deconvolution problem under competing risks

Christophe Chesneau, Isha Dewan, Hassan Doosti*

*Corresponding author for this work

Research output: Contribution to journalArticle

Abstract

Bagai and Prakasa Rao [Analysis of survival data with two dependent competing risks. Biometr J. 1992;7:801–814] considered a competing risks model with two dependent risks. The two risks are initially independent but dependence arises because of the additive effect of an independent risk on the two initially independent risks. They showed that the ratio of failure rates are identifiable in the nonparametric set-up. In this paper, we consider it as a measurement error/deconvolution problem and suggest a nonparametric kernel-type estimator for the ratio of two failure rates. The local error properties of the proposed estimator are studied. Simulation studies show the efficacy of the proposed estimator.

Original languageEnglish
Pages (from-to)331-346
Number of pages16
JournalStatistics
Volume51
Issue number2
DOIs
Publication statusPublished - Mar 2017
Externally publishedYes

Keywords

  • competing risks
  • deconvolution
  • failure rates
  • identifiability

Fingerprint Dive into the research topics of 'On a deconvolution problem under competing risks'. Together they form a unique fingerprint.

Cite this