Population level information combined parameter estimation from complex survey datasets

TL;DR

This paper develops an empirical likelihood method for population-level inference from complex survey data, incorporating informative sampling and auxiliary information to improve estimation efficiency.

Contribution

It introduces a new estimator based on conditional weights that combines survey weights and population information, with proven consistency and asymptotic normality.

Findings

Estimator is strongly consistent and asymptotically normal.

Provides more efficient estimates than existing probability-weighted methods.

Application to demographic hazard modeling demonstrates practical utility.

Abstract

We consider an empirical likelihood framework for inference for a statistical model based on an informative sampling design and population-level information. The population-level information is summarized in the form of estimating equations and incorporated into the inference through additional constraints. Covariate information is incorporated both through the weights and the estimating equations. The estimator is based on conditional weights. We show that under usual conditions, with population size increasing unbounded, the estimates are strongly consistent, asymptotically unbiased, and normally distributed. Moreover, they are more efficient than other probability-weighted analogs. Our framework provides additional justification for inverse probability weighted score estimators in terms of conditional empirical likelihood. We give an application to demographic hazard modeling by combining birth registration data with panel survey data to estimate annual first birth probabilities.