Zheng Q
Division of Biometry and Risk Assessment, National Center for Toxicological Research, Jefferson, Arkansas 72079, USA.
Biometrics. 1998 Mar;54(1):384-8.
The method based on the Kolmogorov backward equations of Little (1995, Biometrics 51, 1278-1291) for computing hazard functions for the multistage carcinogenesis models fails when model parameters are time-dependent. In addition to suggesting an alternative method based on the Kolmogorov forward equation, this note highlights the interplay of the forward equation, the backward equation, and the characteristic method. Advantages and disadvantages of the forward and backward equations are discussed.
基于利特尔（1995年，《生物统计学》第51卷，第1278 - 1291页）的柯尔莫哥洛夫向后方程来计算多阶段致癌模型风险函数的方法，在模型参数随时间变化时会失效。除了提出一种基于柯尔莫哥洛夫向前方程的替代方法外，本注释还强调了向前方程、向后方程和特征线法之间的相互作用。讨论了向前方程和向后方程的优缺点。
