Results of 2018 Best Student Paper Award of the Survey Methods Section
The Survey Methods Section is pleased to announce that Kanika Grover from the University of Manitoba has won the 2018 best student paper award of $300 for her paper co-authored with Elif Acar and Mahmoud Torabi from the University of Manitoba. The winning paper is entitled “Using Copulas in Small Area Estimation.” This award was open to all students who presented at the 2018 annual meeting of the Society in the area of survey methods.
Here is the abstract of the winning paper:
Small area estimation has received considerable attention in recent years due to a growing interest in analyzing domains/areas with very small sample sizes. In order to achieve reliable predictions for such areas, improvised small area techniques need to be adopted. One such technique has been recently proposed using multivariate exchangeable copulas to characterize the error distribution under the linear regression model. While this model offers a highly flexible framework for small area predictions, obtaining reliable estimators of the mean squared prediction error (MSPE) under this model is not straightforward. In addition, the MSPE estimators may perform poorly when such a complex model is misspecified. This paper introduces a likelihood framework to estimate the intra-class dependence of the multivariate exchangeable copula for empirical best unbiased prediction (EBUP) of small area means. We consider both parametric and semi-parametric approaches, and outline a bootstrap method under each approach to obtain a nearly unbiased estimate of the MSPE of the EBUP of small area means. A Monte Carlo simulation study is conducted to evaluate performance of the proposed methods under potential model misspecifications.