Itemwise conditionally independent nonresponse modeling for incomplete\n multivariate data

2016/09/02 by Mauricio Sadinle, Jerome P. Reiter, Sadinle, Mauricio +1 · 1 citation
Mathematics · #Advanced Causal Inference Techniques #FOS: Computer and information sciences #Methodology (stat.ME) #Statistical Methods and Bayesian Inference #Statistical Methods and Inference

paper · pdf · doi:10.48550/arxiv.1609.00656

openalex publication_date 2016/09/02 · openalex created_date 2022/10/01 · openalex updated_date 2026/07/28

Abstract

We introduce a nonresponse mechanism for multivariate missing data in which\neach study variable and its nonresponse indicator are conditionally independent\ngiven the remaining variables and their nonresponse indicators. This is a\nnonignorable missingness mechanism, in that nonresponse for any item can depend\non values of other items that are themselves missing. We show that, under this\nitemwise conditionally independent nonresponse assumption, one can define and\nidentify nonparametric saturated classes of joint multivariate models for the\nstudy variables and their missingness indicators. We also show how to perform\nsensitivity analysis to violations of the conditional independence assumptions\nencoded by this missingness mechanism. Throughout, we illustrate the use of\nthis modeling approach with data analyses.\n

Citations

• Bayesian Density Estimation and Inference Using Mixtures
• Inference and missing data

Cited by

• Markov Missing Graph: A Graphical Approach for Missing Data Imputation

Related

• Sequentially additive nonignorable missing data modeling using auxiliary marginal information
• Sequential identification of nonignorable missing data mechanisms
• Discrete Choice Models for Nonmonotone Nonignorable Missing Data: Identification and Inference
• Graphical Models for Processing Missing Data
• Non-standard conditionally specified models for non-ignorable missing data