Learning from missing data with the binary latent block model

DOI10.1007/S11222-021-10058-YMaRDI QIDQ2066751zbMATH OpenOpenAlexFDO

Authors Gabriel Frisch, Jean-Benoist Leger, Yves Grandvalet

Publication date 14 January 2022

Published in Statistics and Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/2010.12222

zbMATH Keywords

co-clusteringlatent block modelmissing datavariational inferenceMNAR

Mathematics Subject Classification ID

Computational methods for problems pertaining to statistics (62-08) Classification and discrimination; cluster analysis (statistical aspects) (62H30) Missing data (62D10)

Abstract: Missing data can be informative. Ignoring this information can lead to misleading conclusions when the data model does not allow information to be extracted from the missing data. We propose a co-clustering model, based on the Latent Block Model, that aims to take advantage of this nonignorable nonresponses, also known as Missing Not At Random data (MNAR). A variational expectation-maximization algorithm is derived to perform inference and a model selection criterion is presented. We assess the proposed approach on a simulation study, before using our model on the voting records from the lower house of the French Parliament, where our analysis brings out relevant groups of MPs and texts, together with a sensible interpretation of the behavior of non-voters.

Recommendations

• Model-based classification of clustered binary data with non-ignorable missing values
• Diagonal latent block model for binary data
• Model-based co-clustering for mixed type data
• A Latent Variable Model with Non-Ignorable Missing Data
• A latent class selection model for nonignorably missing data

Cites work

• scientific article; zbMATH DE number 2038320 (Why is no real title available?)
• An improvement of the NEC criterion for assessing the number of clusters in a mixture model
• An introduction to variational methods for graphical models
• Block clustering with Bernoulli mixture models: comparison of different approaches
• Block clustering with collapsed latent block models
• Choosing starting values for the EM algorithm for getting the highest likelihood in multivariate Gaussian mixture models
• Co-Clustering of Ordinal Data via Latent Continuous Random Variables and Not Missing at Random Entries
• EM for mixtures
• Estimation and selection for the latent block model on categorical data
• Identifiability of linear mixed effects models
• Inference and missing data
• Latent block model for contingency table
• Model selection for Gaussian latent block clustering with the integrated classification likelihood
• Model-based co-clustering for mixed type data
• Model-based co-clustering for ordinal data
• Overlapping stochastic block models with application to the French political blogosphere
• Spectral clustering and the high-dimensional stochastic blockmodel
• Statistical field theory. With a foreword by David Pines
• Variational Inference for Stochastic Block Models From Sampled Data

Cited in

(2)

• Greedy structure learning from data that contain systematic missing values
• A survey on model-based co-clustering: high dimension and estimation challenges

Describes a project that uses

Uses Software

• PyTorch
• blockcluster

This page was built for publication: Learning from missing data with the binary latent block model

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2066751)