Identifiability of the Sign of Covariate Effects in the Competing Risks Model

Simon M.S. Lo, Ralf Wilke

Research output: Contribution to journalJournal articleResearchpeer-review

216 Downloads (Pure)

Abstract

We present a new framework for the identification of competing risks models, which also include Roy models. We show that by establishing a Hicksian-type decomposition, the direction of covariate effects on the marginal distributions of the competing risks model can be identified under weak restrictions. Our approach leaves the marginal distributions and their joint distribution completely unspecified, except that the associated copula is invariant in the covariates. Results from simulations and two data examples suggest that our method often outperforms existing comparable approaches in terms of the range of durations for which the direction of the covariate effect is identified, particularly for long duration.
Original languageEnglish
JournalEconometric Theory
Volume33
Issue number5
Pages (from-to)1186-1217
Number of pages32
ISSN0266-4666
DOIs
Publication statusPublished - 2017

Bibliographical note

Published online: 3. October 2016

Cite this