Some Prior and Posterior Distributions in Survival Analysis: A Critical Insight on Relationships Derived from Cross-Sectional Data
论文指出，如果生理特征以相加方式影响风险函数，那么幸存者群体中该特征的回归函数会随时间变化，即使个体特征本身不变；并且个体特征与年龄的关系无法从横截面数据唯一估计，除非已知其对风险函数的贡献形式。
Summary Many physiological characteristics such as, for example, blood pressure, level of serum cholesterol, vital capacity, may be considered as ageing factors contributing to the overall hazard rate function in an additive fashion (formula (3.2)). If so, the regression function of the characteristic on time, in a population of survivors to time t, depends on t even if the characteristic, itself, does not vary with t, for any one individual. Also, if there is a relationship between a characteristic and age for an individual, it cannot be uniquely estimated from population cross-sectional data, unless its functional contribution to the hazard rate is determined. Special consideration is given to the analysis, when the prior (i.e. at t = 0) distribution of each characteristic is normal, and the characteristics are independent.