Robust Small-Area Estimation Under Semiparametric Mixed Models

• Activity Number: 634
• Type: Topic Contributed
• Date/Time: Thursday, August 13, 2015 : 8:30 AM to 10:20 AM
• Sponsor: Survey Research Methods Section
• Author(s): Sanjoy Sinha*
• Companies: Carleton College
• Keywords: Bootstrap ; Mean squared prediction error ; Outliers ; Random effects ; Small area mean ; Unit level model

Abstract

Small area estimation has been extensively studied in the literature under unit level linear mixed models. In particular, empirical best linear unbiased predictors (EBLUPs) of small area means and associated estimators of mean squared prediction error (MSPE) have been developed. However, EBLUPs can be sensitive to potential outliers in the data. To deal with outliers, robust EBLUP methods have been developed in the framework of linear mixed models. In this research, we relax the assumption of linear regression for the fixed part of the model and replace it by a weaker assumption of a semi-parametric regression. Approximating the semi-parametric mixed model by a penalized spline mixed model, we develop robust EBLUPs of small area means and bootstrap estimators of MSPEs. We investigate the empirical properties of the proposed estimators using simulations.

(This is a joint research with J. N. K. Rao and L. Dumitrescu, Carleton University)