Empirical Likelihood Inference With Public-Use Survey Data

TL;DR

This paper develops empirical likelihood methods for inference using public-use survey data, providing theoretical foundations, practical implementation guidance, and demonstrating effectiveness through simulations and real data analysis.

Contribution

It introduces pseudo and sample empirical likelihood techniques tailored for complex survey data, with new theoretical results and practical algorithms.

Findings

Methods perform well in simulations

Effective for hypothesis testing with survey data

Applicable to real-world social science datasets

Abstract

Public-use survey data are an important source of information for researchers in social science and health studies to build statistical models and make inferences on the target finite population. This paper presents two general inferential tools through the pseudo empirical likelihood and the sample empirical likelihood methods. Theoretical results on point estimation and linear or nonlinear hypothesis tests involving parameters defined through estimating equations are established, and practical issues with the implementation of the proposed methods are discussed. Results from simulation studies and an application to the 2016 General Social Survey dataset of Statistics Canada show that the proposed methods work well under different scenarios. The inferential procedures and theoretical results presented in the paper make the empirical likelihood a practically useful tool for users of complex survey data.