STATISTICAL METHODS FOR NON STANDARD DATA

这笔赠款的目的是开发有用的统计方法 用于慢性病的流行病学队列和病例对照研究 和传染病。重点将放在发展 (1)有依赖关系的病例对照研究的统计方法 数据；(2)具有二元性状的遗传研究；(3)具有以下特征的队列研究 有关死亡原因及/或日期的部分资料及(4)个案- 流行病例对照研究。 上述四种不同类型的研究有一个共同点 公共的，即以下两个假设中至少有一个是 被侵犯了。它们是(I)观测数据被假定为 相互独立和(Ii)疾病状况，特别是 起因和发生日期，被认为是清楚的 已确认身份。在这笔拨款中，对于四项研究中的每一项，我们将：(A) 指出现有统计数据的位置和程度 方法分解，(B)开发适用于 研究和(C)将这些方法应用于以下数据集 工作的动力来自于。

The objective of this grant is to develop statistical methods useful for epidemiologic cohort and case-control studies of both chronic and infectious diseases. The focus will be on the development of statistical methods for (1) case-control studies with dependent data; (2) genetic studies with binary traits; (3) cohort studies with partial information on causes and/or dates of death and (4) case- control studies with prevalent cases. The above four different types of studies share one thing in common, namely, at least one of the following two assumptions is violated. They are (i) the observed data are assumed to be independent of each other and (ii) the disease status, especially the cause and the date of occurrence, are assumed to be clearly identified. In this grant, for each one of four studies, we will (A) point out the place and the extent to which the existing statistical method breaks down, (b) develop new methods appropriate for the study and (C) apply the methods to the data set for which this work was motivated from.