On the Identifiability of Latent Class Models for Multiple-Systems Estimation

TL;DR

This paper establishes necessary and sufficient conditions for the identifiability of latent class models used in multiple-systems estimation, crucial for human rights data analysis, and offers a verification mechanism for models with individual heterogeneity.

Contribution

It provides a complete characterization of when latent class models are identifiable in multiple-systems estimation, including a new verification method for models with heterogeneity.

Findings

Derived necessary and sufficient conditions for model identifiability.

Developed a mechanism to verify identifiability in heterogeneous models.

Enhanced understanding of latent class model applicability in human rights estimation.

Abstract

Latent class models have recently become popular for multiple-systems estimation in human rights applications. However, it is currently unknown when a given family of latent class models is identifiable in this context. We provide necessary and sufficient conditions on the number of latent classes needed for a family of latent class models to be identifiable. Along the way we provide a mechanism for verifying identifiability in a class of multiple-systems estimation models that allow for individual heterogeneity.