Skip Navigation

Biometrika 2001 88(2):551-564; doi:10.1093/biomet/88.2.551
© 2001 by Biometrika Trust
This Article
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Email this article to a friend
Right arrow Similar articles in this journal
Right arrow Alert me to new issues of the journal
Right arrow Add to My Personal Archive
Right arrow Download to citation manager
Right arrowRequest Permissions
Google Scholar
Right arrow Articles by Ibrahim, J. G.
Right arrow Articles by Lipsitz, S. R.
Right arrow Search for Related Content
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

Missing responses in generalised linear mixed models when the missing data mechanism is nonignorable

Joseph G.Ibrahim1, Ming-Hui Chen2 and Stuart R.Lipsitz3

1 Department of Biostatistics, Harvard School of Public Health and Dana-Farber Cancer Institute, 44 Binney Street, Boston, Massachusetts 02115, U.S.Aibrahim{at}jimmy.harvard.edu 2 Department of Mathematical Sciences, Worcester Polytechnic Institute, 100 Institute Road, Worcester, Massachusetts 01609, U.S.A.mhchen{at}wpi.edu 3 Department of Biostatistics and Epidemiology, Medical University of South Carolina, 135 Rutledge Avenue, Charleston, South Carolina 29425, U.S.A. lipsitzs{at}musc.edu

We propose a method for estimating parameters in the generalised linear mixed model with nonignorable missing response data and with nonmonotone patterns of missing data in the response variable. We develop a Monte Carlo EM algorithm for estimating the parameters in the model via the Gibbs sampler.For the normal random effects model, we derive a novel analytical form for the E- and M-steps, which is facilitated by integrating out the random effects. This form leads to a computationally feasible and extremely efficient Monte Carlo EM algorithm for computing maximum likelihood estimates and standard errors. In addition, we propose a very general joint multinomial model for the missing data indicators, which can be specified via a sequence of one-dimensional conditional distributions. This multinomial model allows for an arbitrary correlation structure between the missing data indicators, and has the potential of reducing the number of nuisance parameters. Real datasets from the International Breast Cancer Study Group and an environmental study involving dyspnoea in cotton workers are presented to illustrate the proposed methods.

Key Words: EM algorithm; Gibbs sampling; Maximum likelihood estimation; Missing data mechanism; Monte Carlo EM; Random effects model

Received November 1999. Revised August 2000


Add to CiteULike CiteULike   Add to Connotea Connotea   Add to Del.icio.us Del.icio.us    What's this?


This article has been cited by other articles:


Home page
 BiometrikaHome page
S. Vansteelandt, A. Rotnitzky, and J. Robins
Estimation of Regression Models for the Mean of Repeated Outcomes Under Nonignorable Nonmonotone Nonresponse
Biometrika, December 1, 2007; 94(4): 841 - 860.
[Abstract] [PDF]

Home page
 BiostatisticsHome page
M. Geraci and M. Bottai
Quantile regression for longitudinal data using the asymmetric Laplace distribution
Biostat., January 1, 2007; 8(1): 140 - 154.
[Abstract] [Full Text] [PDF]

Home page
 Statistical ModellingHome page
M. Geraci and M. Bottai
Use of auxiliary data in semi-parametric spatial regression with nonignorable missing responses
Statistical Modeling, December 1, 2006; 6(4): 321 - 336.
[Abstract] [PDF]


Disclaimer: Please note that abstracts for content published before 1996 were created through digital scanning and may therefore not exactly replicate the text of the original print issues. All efforts have been made to ensure accuracy, but the Publisher will not be held responsible for any remaining inaccuracies. If you require any further clarification, please contact our Customer Services Department.