DOI: 10.2307/2531021
关键词: Sample size determination 、 Statistics 、 Demography 、 Population 、 Patient population 、 Proportional hazards regression 、 Survival analysis 、 Mathematics 、 Concomitant 、 Regression analysis 、 Asymptotic power
摘要: A formula is derived for determining the number of observations necessary to test equality two survival distributions when concomitant information incorporated. This should be useful in designing clinical trials with a heterogeneous patient population. Schoenfeld (1981, Biometrika 68, 316-319) asymptotic power class statistics used distributions. That result extended case where available each individual and proportional-hazards model holds. The loss efficiency caused by ignoring variables also computed.