Title: Nonparametric estimation of repeated densities with heterogeneous sample sizes
Authors: Jiaming Qiu - Iowa State University (United States)
Xiongtao Dai - Iowa State University (United States) [presenting]
Zhengyuan Zhu - Iowa State University (United States)
Abstract: The estimation of densities in multiple subpopulations is considered, where the available sample size in each subpopulation greatly varies. For example, in epidemiology, different diseases may share similar pathogenic mechanism but differ in their prevalence. Without specifying a parametric form, the proposed approach pools information from the population and estimate the density in each subpopulation in a data-driven fashion. Low-dimensional approximating density families in the form of exponential families are constructed from the principal modes of variation in the log-densities, within which subpopulation densities are then fitted based on likelihood principles and shrinkage. The proposed methods are shown to be interpretable and efficient in simulation as well as applications to electronic medical record and rainfall data.