COHORT SIZE AND MAXIMUM-LIKELIHOOD-ESTIMATION OF MORTALITY PARAMETERS

The purpose of this study was to investigate the effects of cohort siz e on maximum likelihood estimates of mortality parameters. Recent expe rimental investigations have stressed the importance of large cohorts for detecting leveling off of mortality rates at older ages. In the pr esent study, emphasis was placed on evaluation of relatively small coh orts (about 150-300 individuals). Deaths were simulated under the assu mption of the frailty mortality model. Two different parameter sets th at resulted in differences in mean life span of more than twofold were used for simulations. Our smallest cohorts yielded parameter estimate s that had generally good statistical properties, but relatively large standard errors. For tests of hypotheses concerning equality of param eters among populations or experimental treatments, empirical standard errors (obtained from several cohorts) were preferable to asymptotic standard errors (obtained for single cohorts). In particular, empirica l standard en ors yielded reliable type I error rates. (C) 1998 Elsevi er Science Inc.