Block-conditional missing at random models for missing data

DOI10.1214/10-STS344MaRDI QIDQ906522zbMATHFDO

Authors Yan Zhou, Roderick Little, John D. Kalbfleisch

Publication date 22 January 2016

Published in Statistical Science (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1104.2400

zbMATH Keywords

EM algorithmcategorical datablock-conditional MAR modelsblock-sequential missing data models

Mathematics Subject Classification ID

Point estimation (62F10) Classification and discrimination; cluster analysis (statistical aspects) (62H30)

Abstract: Two major ideas in the analysis of missing data are (a) the EM algorithm [Dempster, Laird and Rubin, J. Roy. Statist. Soc. Ser. B 39 (1977) 1--38] for maximum likelihood (ML) estimation, and (b) the formulation of models for the joint distribution of the data $Z$ and missing data indicators $M$, and associated "missing at random"; (MAR) condition under which a model for $M$ is unnecessary [Rubin, Biometrika 63 (1976) 581--592]. Most previous work has treated $Z$ and $M$ as single blocks, yielding selection or pattern-mixture models depending on how their joint distribution is factorized. This paper explores "block-sequential"; models that interleave subsets of the variables and their missing data indicators, and then make parameter restrictions based on assumptions in each block. These include models that are not MAR. We examine a subclass of block-sequential models we call block-conditional MAR (BCMAR) models, and an associated block-monotone reduced likelihood strategy that typically yields consistent estimates by selectively discarding some data. Alternatively, full ML estimation can often be achieved via the EM algorithm. We examine in some detail BCMAR models for the case of two multinomially distributed categorical variables, and a two block structure where the first block is categorical and the second block arises from a (possibly multivariate) exponential family distribution.

Recommendations

• Monotone missing data and pattern‐mixture models
• Missing at random, likelihood ignorability and model completeness.
• A class of pattern-mixture models for normal incomplete data
• Likelihood based frequentist inference when data are missing at random
• Lattice-ordered conditional independence models for missing data

Cites work

• scientific article; zbMATH DE number 3519741 (Why is no real title available?)
• scientific article; zbMATH DE number 3567782 (Why is no real title available?)
• scientific article; zbMATH DE number 2140075 (Why is no real title available?)
• scientific article; zbMATH DE number 1834445 (Why is no real title available?)
• A Pattern-Mixture Model for Longitudinal Binary Responses with Nonignorable Nonresponse
• A Weighted Estimating Equation for Missing Covariate Data with Properties Similar to Maximum Likelihood
• Addressing complications of intention-to-treat analysis in the combined presence of all-or-none treatment-noncompliance and subsequent missing outcomes
• An Extended General Location Model for Causal Inferences from Data Subject to Noncompliance and Missing Values
• Fusing multimodal biometrics with quality estimates via a Bayesian belief network
• Inference and missing data
• Maximum Likelihood Estimation from Incomplete Data
• Maximum likelihood estimation via the ECM algorithm: A general framework
• Modeling the Drop-Out Mechanism in Repeated-Measures Studies
• Partial likelihood
• Pattern-Mixture Models for Multivariate Incomplete Data
• Subsample Ignorable Likelihood for Regression Analysis with Missing Data
• Two-Dimensional Contingency Tables with Both Completely and Partially Cross-Classified Data

Cited in

(7)

• A self-censoring model for multivariate nonignorable nonmonotone missing data
• Semiparametric Inference for Nonmonotone Missing-Not-at-Random Data: The No Self-Censoring Model
• Conditional moment models with data missing at random
• Imputations for High Missing Rate Data in Covariates Via Semi-supervised Learning Approach
• Analysis of local sensitivity to nonignorability with missing outcomes and predictors
• Global sensitivity analysis of randomized trials with nonmonotone missing binary outcomes: Application to studies of substance use disorders
• Assessable and interpretable sensitivity analysis in the pattern graph framework for nonignorable missingness mechanisms

This page was built for publication: Block-conditional missing at random models for missing data

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q906522)