Abstract:We consider modeling, inference, and computation for analyzing multivariate binary data. We propose a new model that consists of a low dimensional latent variable component and a sparse graphical component. Our study is motivated by analysis of item response data in cognitive assessment and has applications to many disciplines where item response data are collected. Standard approaches to item response data in cognitive assessment adopt the multidimensional item response theory (IRT) models. However, human cognition is typically a complicated process and thus may not be adequately described by just a few factors. Consequently, a low-dimensional latent factor model, such as the multidimensional IRT models, is often insufficient to capture the structure of the data. The proposed model adds a sparse graphical component that captures the remaining ad hoc dependence. It reduces to a multidimensional IRT model when the graphical component becomes degenerate. Model selection and parameter estimation are carried out simultaneously through construction of a pseudo-likelihood function and properly chosen penalty terms. The convexity of the pseudo-likelihood function allows us to develop an efficient algorithm, while the penalty terms generate a low-dimensional latent component and a sparse graphical structure. Desirable theoretical properties are established under suitable regularity conditions. The method is applied to the revised Eysenck's personality questionnaire, revealing its usefulness in item analysis. Simulation results are reported that show the new method works well in practical situations.

Mixed and missing data: a unified treatment with latent graphical models

High Dimensional Semiparametric Latent Graphical Model for Mixed Data

High-dimensional Mixed Graphical Models

A Novel Bayesian Approach for Latent Variable Modeling from Mixed Data with Missing Values

Graphical model selection with latent variables

Mixed Graphical Models with Missing Data and the Partial Imputation EM Algorithm

High-Dimensional Undirected Graphical Models for Arbitrary Mixed Data

Probabilistic Missing Value Imputation for Mixed Categorical and Ordered Data

Generalized Linear Latent Models for Multivariate Longitudinal Measurements Mixed with Hidden Markov Models.

A Copula-based Imputation Model for Missing Data of Mixed Type in Multilevel Data Sets

Learning Latent Variable Gaussian Graphical Models

Mixed Likelihood Gaussian Process Latent Variable Model

A Fused Latent and Graphical Model for Multivariate Binary Data

Latent Gaussian Graphical Models with Golazo Penalty

Bayesian multiple Gaussian graphical models for multilevel variables from unknown classes

Mixed data Deep Gaussian Mixture Model: A clustering model for mixed datasets

A Generalized Latent Factor Model Approach to Mixed-data Matrix Completion with Entrywise Consistency

Heterogeneous latent transfer learning in Gaussian graphical models

Multilevel latent Gaussian process model for mixed discrete and continuous multivariate response data

Finding mixed memberships in categorical data

Fast estimation of generalized linear latent variable models for performance and process data with ordinal, continuous, and count observed variables