SEMINAR - February 28, 2025

Abstract

Empirical best prediction (EBP) is a well-known method for producing reliable proportion estimates when the primary data source provides only small or no sample from finite populations. There are potential challenges in implementing existing EBP methodology such as limited auxiliary variables in the frame (not adequate for building a reasonable working predictive model) or unable to accurately link the sample to the finite population frame due to absence of identifiers. In this paper, we propose a new data linkage approach where the finite population frame is replaced by a big probability sample, having a large set of auxiliary variables but not the outcome binary variable of interest. We fit an assumed model on the small probability sample and then impute the outcome variable for all units of the big sample to obtain standard weighted proportions. We develop a new adjusted maximum likelihood (ML) method so that the estimate of model variance doesn't fall on the boundary, which is otherwise encountered in commonly used ML method.  We also propose an estimator of the mean squared prediction error using a parametric bootstrap method and address computational issues by developing an efficient Expectation Maximization algorithm. The proposed methodology is illustrated in the context of election projection for small areas.  My talk will be based on a joint paper with Ms. Aditi Sen, an UMD doctoral student.

About the Speaker

Dr. Partha Lahiri is a professor of Survey Methodology and Mathematics at the University of Maryland College Park and an adjunct research professor at the Institute of Social Research, University of Michigan, Ann Arbor.  He is currently serving as the Director of Joint Program in Survey Methodology (JPSM). His areas of research interest include data linkages, Bayesian statistics, survey sampling and small-area estimation.  Dr. Lahiri served on the editorial board of several international journals and on many advisory committees, including the U.S. Census Advisory committee and U.S. National Academy panel.  He also served as an advisor or consultant for various international organizations such as the United Nations and the World Bank.  He is a Fellow of the American Statistical Association and the Institute of Mathematical Statistics and an elected member of the International Statistical Institute. Dr. Lahiri is the recipient of the 2020 SAE award for his outstanding contribution to the research, application, and education of small area estimation.  He was awarded the Neyman Medal in a joint session of the 3rd Congress of Polish Statistics and 2022 International Association of Official Statistics (IAOS) held in Krakow, Poland, for his outstanding contributions to the development of statistical sciences.  Dr. Lahiri is now serving as the President-Elect of the International Association of Survey Statisticians.

Event Organizer

David Kepplinger
Assistant Professor, Department of Statistics
College of Engineering and Computing
George Mason University