本文提出了生命统计资料分析中的年龄-时期、年龄-队列和年龄-线性趋势三个概率模型。这些模型可定量估计发病率或死亡率的年龄效应、时期效应和出生队列效应以及组间的相对危险度;并可提取到“纯”年龄别发病或死亡率。应用这些模型对女性肺癌死亡率和乳腺癌发病率资料进行拟合,结果表明是十分有效的。 This paper presents three probabilistic models of age-period, age-cohort, and age-linear trends in life statistics analysis. These models can quantitatively estimate the age-effect, epoch-effect and birth cohort effects of morbidity or mortality and the relative risk between groups; and can obtain “neat” age-specific morbidity or mortality. Using these models to fit female lung cancer mortality and breast cancer incidence data, the results show that it is very effective.