We consider a model with competing risks failure times and partially
identified interval data. The data problems imply that effects on
risk-specific failure distributions can only be bounded. We develop a
non-parametric bounds analysis of risk-specific cumulative incidence
curves (CIC) to bound a difference-in-differences treatment effect on
the CIC over different definitions of the latent durations. Our
simulations demonstrate the applicability of this approach also in case
of dependent competing risks. We then apply our framework to empirically
evaluate the effect of unemployment benefits on observed migration
probabilities in Germany. Our findings weakly indicate that reducing the
maximum receipt of unemployment benefits increases the migration
probability, at least among high-skilled individuals.
“) ;
$p->section(“Paper Download”, “Full Text (PDF)
Presented by:
Ralf Wilke (University of Leicester)
Date & time:
March 19, 2007 4:00 pm - March 19, 2007 12:00 am
External seminars home