Fitting multilevel multivariate models with missing data in responses and covariates that may include interactions and non-linear terms

Harvey Goldstein*, James R. Carpenter, William J. Browne

*Corresponding author for this work

Research output: Contribution to journalArticle (Academic Journal)peer-review

48 Citations (Scopus)
424 Downloads (Pure)

Abstract

The paper extends existing models for multilevel multivariate data with mixed response types to handle quite general types and patterns of missing data values in a wide range of multilevel generalized linear models. It proposes an efficient Bayesian modelling approach that allows missing values in covariates, including models where there are interactions or other functions of covariates such as polynomials. The procedure can also be used to produce multiply imputed complete data sets. A simulation study is presented as well as the analysis of a longitudinal data set. The paper also shows how existing multiprocess models for handling endogeneity can be extended by the framework proposed.

Original languageEnglish
Pages (from-to)553-564
Number of pages12
JournalJournal of the Royal Statistical Society: Series A
Volume177
Issue number2
DOIs
Publication statusPublished - 6 Feb 2014

Structured keywords

  • WUN
  • Endogeneity; Latent normal model; Markov chain Monte Carlo methods; Missing data; Multilevel modelling; Multiple imputation; Multiprocess model; Multivariate modelling
  • Jean Golding

Keywords

  • Endogeneity
  • Latent normal model
  • Markov chain Monte Carlo methods
  • Missing data
  • Multilevel modelling
  • Multiple imputation
  • Multiprocess model
  • Multivariate modelling
  • MULTIPLE-IMPUTATION
  • Lemma 3

    Rintoul, D. A.

    1/10/1130/09/14

    Project: Research

Cite this