报告题目：Flexible Modeling of Medical Cost Data

报 告 人：刘磊教授（美国西北大学）

报告时间：2015年7月9日（周4）下午3:00-4:00

报告地点：知新楼B-1248

摘要：

Medical cost data are often skewed to theright and heteroscedastic, having a nonlinear relation with covariates. Totackle these issues, we consider an extension to generalized linear models byassuming nonlinear covariate effects in the mean function and allowing thevariance to be an unknown but smooth function of the mean. We make no furtherassumption on the distributional form. The unknown functions are described bypenalized splines, and the estimation is carried out using nonparametricquasi-likelihood. Simulation studies show the flexibility and advantages of ourapproach. We apply the model to the annual medical costs of heart failurepatients in the clinical data repository (CDR) at the University of VirginiaHospital System. We also discuss how to adopt this modeling framework incorrelated medical costs data.

热烈欢迎各位老师和同学参加！