Modeling data with informative cluster size

具有信息丰富的簇大小的数据建模

• 批准号: 8351257
• 负责人: Zhen Chen
• 金额: $ 7万
• 依托单位: EUNICE KENNEDY SHRIVER NATIONAL INSTITUTE OF CHILD HEALTH & HUMAN DEVELOPMENT
• 依托单位国家: 美国
• 资助国家: 美国
• 起止时间: 至
• 项目状态: 未结题
• 来源: https://reporter.nih.gov/project-details/8351257
• 关键词: Data; Epidemiologic Studies; Joints; Literature; Methods; Modeling; Outcome; Procedures; Sampling; Statistical Methods; Weight; data modeling; follow-up; novel strategies

When data are clustered due to longitudinal follow up or repeated sampling, special statistical methods need to be used to avoid bias and inefficiency. While in some clustered data the cluster size is pre-determined, in others it may be varying and correlated to the outcome of subunits, resulting in informative cluster size. When the cluster size is informative, standard statistical procedures that ignore cluster size may produce biased estimates. In this PI-initiated project, we will compare several methods that have been proposed in the literature to model data with informative cluster size, including within-cluster resampling, cluster weighted GEE, and joint modeling.

当由于纵向跟踪或重复抽样而对数据进行聚类时，需要使用特殊的统计方法来避免偏差和低效率。虽然在一些聚类数据中，聚类大小是预先确定的，但在其他数据中，它可能是变化的，并与子单元的结果相关，从而产生信息聚类大小。当集群大小具有信息量时，忽略集群大小的标准统计程序可能产生有偏差的估计。在这个pi发起的项目中，我们将比较文献中提出的几种方法，包括聚类内重采样、聚类加权GEE和联合建模，以提供信息簇大小的数据建模。