A Unified Bayesian Framework for Modeling Measurement Error in Multinomial Data

TL;DR

This paper introduces a Bayesian hierarchical framework that simultaneously models false negatives and false positives in multinomial data, improving inference accuracy across diverse fields.

Contribution

It presents a unified Bayesian approach to handle both types of measurement error in multinomial data, addressing limitations of existing methods.

Findings

The method performs well on simulated data.

Application to acoustic bat monitoring data.

Application to official crime data.

Abstract

Measurement error in multinomial data is a well-known and well-studied inferential problem that is encountered in many fields, including engineering, biomedical and omics research, ecology, finance, official statistics, and social sciences. Methods developed to accommodate measurement error in multinomial data are typically equipped to handle false negatives or false positives, but not both. We provide a unified framework for accommodating both forms of measurement error using a Bayesian hierarchical approach. We demonstrate the proposed method's performance on simulated data and apply it to acoustic bat monitoring and official crime data.