References & Citations
Statistics > Methodology
Title: Identifiability of Latent Class Models with Covariates
(Submitted on 27 Mar 2021 (v1), last revised 8 Feb 2022 (this version, v3))
Abstract: Latent class models with covariates are widely used for psychological, social, and educational research. Yet the fundamental identifiability issue of these models has not been fully addressed. Among the previous research on the identifiability of latent class models with covariates, Huang and Bandeen-Roche (2004, Psychometrika, 69:5-32) studied the local identifiability conditions. However, motivated by recent advances in the identifiability of the restricted latent class models, particularly Cognitive Diagnosis Models (CDMs), we show in this work that the conditions in Huang and Bandeen-Roche (2004) are only necessary but not sufficient to determine the local identifiability of the model parameters. To address the open identifiability issue for latent class models with covariates, this work establishes conditions to ensure the global identifiability of the model parameters in both strict and generic senses. Moreover, our results extend to the polytomous-response CDMs with covariates, which generalizes the existing identifiability results for CDMs.
Submission history
From: Jing Ouyang [view email][v1] Sat, 27 Mar 2021 12:04:05 GMT (38kb)
[v2] Tue, 6 Apr 2021 22:38:03 GMT (38kb)
[v3] Tue, 8 Feb 2022 02:00:20 GMT (39kb)
Link back to: arXiv, form interface, contact.