This paper develops a consistent test for the correct hazard rate specification within the context of random right hand censoring of the dependent variable. The test is based on comparing a parametric estimate with a kernel estimate of the hazard rate. We establish the asymptotic distribution of the test statistic under the null hypothesis of correct… (More)
The authors would like to thank three anonymous referees and the editor for their comments. Any remaining errors are the responsibility of the authors. 2 Abstract The usual standard errors for the regression coeecients in a Seemingly Unrelated Regression model have a substantial downward bias. Bootstrapping the standard errors does not seem to improve… (More)
This paper considers nonparametric identification of " latent " competing risks and Roy duration models in which one does not know which process has been observed. It is shown that these models are identifiable without the usual conditional independence and exclusion restrictions.