A1588
Title: On the implications of proportional hazards assumptions for competing risks modelling
Authors: Ralf Wilke - Copenhagen Business School (Denmark) [presenting]
Ming Sum Simon Lo - United Arab Emirates University (United Arab Emirates)
Takeshi Emura - Hiroshima University (Japan)
Abstract: The assumption of hazard rates being proportional in covariates is widely made in empirical research and extensive research has been done to develop tests of its validity. Rather than contributing to this literature, new insights are provided on the implications of proportional hazards (PH) modelling in competing risks models. It is shown that the use of a PH model for the cause-specific hazards or subdistribution hazards can strongly restrict the class of copulas and marginal hazards compatible with a competing risks model. Empirical researchers should be aware that working with these models can be so restrictive that only degenerate or independent risks models are compatible. Numerical results confirm that estimates from cause-specific hazards models are not informative about patterns in the data generating process.
