学术报告

题      目： A general averaging method for count data with overdispersion and/or excess zeros in biomedicine

报  告  人：刘寅   副教授  (邀请人：张旭 )

                                     中南财经政法大学统计与数学学院

时      间：4月21日  14：00-15：00

地     点：数科院东楼401

报告人简介:

       刘寅，中南财经政法大学统计与数学学院副教授，硕士生导师。主要研究方向是应用统计及数理统计，近年来主要聚焦于复杂数据建模分析、模型平均及统计计算。

摘      要：

       A new optimal weighting based on cross-validation method is developed for count data with overdispersion and/or excess zeros. The method includes the zero-inflated negative binomial model. A K-fold cross-validation technique is used to select the optimal weight vector. The method enhances computational efficiency by deleting one group of observations instead of one observation. The proposed method is superior to three commonly used methods on simulation studies and empirical applications.