报告时间：2017年4月21日上午9:00-10:00

报告地点：数计学院4号楼225

报告题目(Title):  Efficient Small Area Estimation with Spatio-temporal Fay-Herriot Models When Covariates Are Measured With Error

摘要(Abstract):  Small area estimation methods typically combine direct estimates from a survey and predictions from a model in order to obtain estimates of population quantities such that the mean squared error of the predictor is minimized. When covariates in the Spatio-temporal Fay-Herriot model are measured with error, we propose an strategy to improve the predictor of the parameter of interest in the small area and incease its effciency by considering an additive measurement error model and using the well-known bias correction technique called simulation-extrapolation. Our proposed predictor is shown to perform better empirically than the predictor proposed by Marhuenda et al. (2013).