CMStatistics 2021: Start Registration
View Submission - CMStatistics
Title: A nonparametric instrumental approach to endogeneity in competing risks models Authors:  Jad Beyhum - KU Leuven (Belgium) [presenting]
Jean-Pierre Florens - Toulouse School of Economics (France)
Ingrid Van Keilegom - KU Leuven (Belgium)
Abstract: Treatment models with duration outcomes, competing risks and random right censoring are discussed. The endogeneity issue is solved using a discrete instrumental variable. We show that the competing risks model generates a non-parametric quantile instrumental regression problem. The cause-specific cumulative incidence, the cause-specific hazard and the subdistribution hazard can be recovered from the regression function. A distinguishing feature of the model is that censoring and competing risks prevent identification at some quantiles. We characterize the set of quantiles for which exact identification is possible and give partial identification results for other quantiles. We outline an estimation procedure and discuss its properties. The finite sample performance of the estimator is evaluated through simulations. We apply the proposed method to the Health Insurance Plan of Greater New York experiment.