Open AccessPosted Content
Maximum Likelihood Estimation in Latent Class Models For Contingency Table Data
Reads0
Chats0
TLDR
In this article, the basic latent class model proposed originally by the sociologist Paul F. Lazarfeld for categorical variables is studied and its geometric structure is explained. And the authors draw parallels between the statistical and geometric properties of latent class models and illustrate geometrically the causes of many problems associated with maximum likelihood estimation and related statistical inference.Abstract:
Statistical models with latent structure have a history going back to the 1950s and have seen widespread use in the social sciences and, more recently, in computational biology and in machine learning. Here we study the basic latent class model proposed originally by the sociologist Paul F. Lazarfeld for categorical variables, and we explain its geometric structure. We draw parallels between the statistical and geometric properties of latent class models and we illustrate geometrically the causes of many problems associated with maximum likelihood estimation and related statistical inference. In particular, we focus on issues of non-identifiability and determination of the model dimension, of maximization of the likelihood function and on the effect of symmetric data. We illustrate these phenomena with a variety of synthetic and real-life tables, of different dimension and complexity. Much of the motivation for this work stems from the “100 Swiss Francs” problem, which we introduce and describe in detail.read more
Citations
More filters
Journal ArticleDOI
Copula Gaussian graphical models and their application to modeling functional disability data
Adrian Dobra,Alex Lenkoski +1 more
TL;DR: A comprehensive Bayesian approach for graphical model determination in observational studies that can accommodate binary, ordinal or continuous variables simultaneously simultaneously is proposed.
Book
Algebraic Statistics
TL;DR: This tutorial will consist of a detailed study of two examples where the algebra/statistics connection has proven especially useful: in the study of phylogenetic models and in the analysis of contingency tables.
Journal ArticleDOI
Mixture of latent trait analyzers for model-based clustering of categorical data
TL;DR: A variational approach for fitting the mixture of latent trait models is developed and it is shown to yield intuitive clustering results and it gives a much better fit than either latent class analysis or latent trait analysis alone.
Proceedings ArticleDOI
Maximum likelihood for matrices with rank constraints
TL;DR: This work discusses methods for finding the global maximum of the likelihood function, presents a duality theorem due to Draisma and Rodriguez, and shares recent work with Kubkas and Robeva concerning nonnegative rank and the EM algorithm.
Journal ArticleDOI
Fixed points of the EM algorithm and nonnegative rank boundaries
TL;DR: The set of fixed points of the Expectation-Maximization algorithm is characterized, and the boundary of the space of matrices with nonnegative rank at most $3$ is studied, which correspond to algebraic varieties with many irreducible components.
References
More filters
BookDOI
Ideals, Varieties, and Algorithms
TL;DR: In the Groebner package, the most commonly used commands are NormalForm, for doing the division algorithm, and Basis, for computing a Groebners basis as mentioned in this paper. But these commands require a large number of variables.
Journal ArticleDOI
Mixture densities, maximum likelihood, and the EM algorithm
TL;DR: This work discusses the formulation and theoretical and practical properties of the EM algorithm, a specialization to the mixture density context of a general algorithm used to approximate maximum-likelihood estimates for incomplete data problems.
Book
Ideals, Varieties, and Algorithms: An Introduction to Computational Algebraic Geometry and Commutative Algebra
TL;DR: Schenzel as mentioned in this paper provides a good introduction to algebraic geometry and commutative algebra with a strong perspective toward practical and computational aspects, including the elimination theorem, the extension theorem, closure theorem, and the Nullstellensatz.
Journal ArticleDOI
Latent Structure Analysis.
Book
Probabilistic Networks and Expert Systems
Robert G. Cowell,Steffen L. Lauritzen,A. Philip David,David Spiegelhalter,V. Nair,J. Lawless,Michael I. Jordan +6 more
TL;DR: This book gives a thorough and rigorous mathematical treatment of the underlying ideas, structures, and algorithms of probabilistic expert systems, emphasizing those cases in which exact answers are obtainable.