Maximum Likelihood Methods for Nonignorable Missing Responses and Covariates in Random Effects Models
- 11 December 2003
- journal article
- Published by Oxford University Press (OUP) in Biometrics
- Vol. 59 (4) , 1140-1150
Abstract
Summary. This article analyzes quality of life (QOL) data from an Eastern Cooperative Oncology Group (ECOG) melanoma trial that compared treatment with ganglioside vaccination to treatment with high‐dose interferon. The analysis of this data set is challenging due to several difficulties, namely, nonignorable missing longitudinal responses and baseline covariates. Hence, we propose a selection model for estimating parameters in the normal random effects model with nonignorable missing responses and covariates. Parameters are estimated via maximum likelihood using the Gibbs sampler and a Monte Carlo expectation maximization (EM) algorithm. Standard errors are calculated using the bootstrap. The method allows for nonmonotone patterns of missing data in both the response variable and the covariates. We model the missing data mechanism and the missing covariate distribution via a sequence of one‐dimensional conditional distributions, allowing the missing covariates to be either categorical or continuous, as well as time‐varying. We apply the proposed approach to the ECOG quality‐of‐life data and conduct a small simulation study evaluating the performance of the maximum likelihood estimates. Our results indicate that a patient treated with the vaccine has a higher QOL score on average at a given time point than a patient treated with high‐dose interferon.Keywords
This publication has 19 references indexed in Scilit:
- Analysis of Multivariate Longitudinal Outcomes With Nonignorable Dropouts and Missing CovariatesJournal of the American Statistical Association, 2002
- Missing responses in generalised linear mixed models when the missing data mechanism is nonignorableBiometrika, 2001
- Semiparametric Regression for Repeated Outcomes with Nonignorable NonresponseJournal of the American Statistical Association, 1998
- The analysis of longitudinal ordinal data with nonrandom drop-outBiometrika, 1997
- A conditional model for incomplete covariates in parametric regression modelsBiometrika, 1996
- Multivariate Logistic Models for Incomplete Binary ResponsesJournal of the American Statistical Association, 1996
- Pattern-Mixture Models for Multivariate Incomplete DataJournal of the American Statistical Association, 1993
- Inference from Iterative Simulation Using Multiple SequencesStatistical Science, 1992
- Methods for the analysis of informatively censored longitudinal dataStatistics in Medicine, 1992
- A Monte Carlo Implementation of the EM Algorithm and the Poor Man's Data Augmentation AlgorithmsJournal of the American Statistical Association, 1990