Maximum likelihood estimation of the latent class model through model boundary decomposition

DOI10.18409/JAS.V10I1.75MaRDI QIDQ5226940zbMATH OpenOpenAlexWikidataFDO

Authors

Publication date 2 August 2019

Published in Journal of Algebraic Statistics (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1710.01696

zbMATH Keywords

expectation maximizationmaximum likelihood estimationlatent class modelsboundary stratificationfixed point ideals

Mathematics Subject Classification ID

Contingency tables (62H17) Multilinear algebra, tensor calculus (15A69) Applications of commutative algebra (e.g., to statistics, control theory, optimization, etc.) (13P25) Semialgebraic sets and related spaces (14P10)

Abstract: The Expectation-Maximization (EM) algorithm is routinely used for the maximum likelihood estimation in the latent class analysis. However, the EM algorithm comes with no guarantees of reaching the global optimum. We study the geometry of the latent class model in order to understand the behavior of the maximum likelihood estimator. In particular, we characterize the boundary stratification of the binary latent class model with a binary hidden variable. For small models, such as for three binary observed variables, we show that this stratification allows exact computation of the maximum likelihood estimator. In this case we use simulations to study the maximum likelihood estimation attraction basins of the various strata. Our theoretical study is complemented with a careful analysis of the EM fixed point ideal which provides an alternative method of studying the boundary stratification and maximizing the likelihood function. In particular, we compute the minimal primes of this ideal in the case of a binary latent class model with a binary or ternary hidden random variable.

Recommendations

• Simple Computation of Maximum Likelihood Estimates in Latent Class Model with Equality and Constant Constraints
• Global convergence of the EM algorithm for unconstrained latent variable models with categorical indicators
• Avoiding Boundary Estimates in Latent Class Analysis by Bayesian Posterior Mode Estimation
• scientific article; zbMATH DE number 4145204
• Parameter estimation in latent profile models.

Cited in

(10)

• Total positivity in exponential families with application to binary variables
• Optimal Transport to a Variety
• Avoiding Boundary Estimates in Latent Class Analysis by Bayesian Posterior Mode Estimation
• Female athletic participation and income: evidence from a latent class model
• Geometry of linear convolutional networks
• A maximum likelihood estimator for quartets under the Cavender-Farris-Neyman model
• Maximum likelihood estimation for unrooted 3-leaf trees: an analytic solution for the CFN model
• New directions in algebraic statistics: three challenges from 2023
• Mixtures of discrete decomposable graphical models
• Maximum likelihood estimation for discrete latent variable models via evolutionary algorithms

This page was built for publication: Maximum likelihood estimation of the latent class model through model boundary decomposition

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q5226940)