Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
This paper describes simple methods for testing various hypotheses of proportionality between the cause-specific hazards in competing risks models. It provides methods for estimating competing risks models with equality restrictions across the cause-specific hazards as single risk models. Thus, likelihood-ratio test statistics of these hypotheses can be conducted using only single risk programs. The first method described involves simple adjustment factors requiring only sample exit proportions. The second involves duplicating the sample and augmenting single risk models with dummy variables. Copyright 1991 by Blackwell Publishing Ltd