Composite linear models for incomplete multinomial data
- 15 March 1994
- journal article
- research article
- Published by Wiley in Statistics in Medicine
- Vol. 13 (5-7) , 609-622
- https://doi.org/10.1002/sim.4780130522
Abstract
A composite linear model (CLM) is a matrix model for incomplete multinomial data. A CLM provides a unified approach for maximum likelihood inference which is applicable to a wide variety of problems involving incomplete multinomial data. By formulating a model as a CLM, one can simplify computation of maximum likelihood estimates and asymptotic standard errors. As an example, we use CLM to test marginal homogeneity for ordered categories, subject to both ignorable and non‐ignorable missing‐data mechanisms.Keywords
This publication has 27 references indexed in Scilit:
- Closed‐form estimates for missing counts in two‐way contingency tablesStatistics in Medicine, 1992
- Finding the design matrix for the marginal homogeneity modelBiometrika, 1990
- Generalizing Logistic Regression by Nonparametric MixingJournal of the American Statistical Association, 1989
- Regression Analysis for Categorical Variables with Outcome Subject to Nonignorable NonresponseJournal of the American Statistical Association, 1988
- Maximum likelihood estimation and likelihood ratio test for square tables with missing dataStatistics in Medicine, 1987
- Causal Models for Patterns of NonresponseJournal of the American Statistical Association, 1986
- Log-Linear Models for Doubly Sampled Categorical Data Fitted by the EM AlgorithmJournal of the American Statistical Association, 1985
- Testing Marginal Homogeneity for Ordinal Categorical VariablesPublished by JSTOR ,1983
- On the Convergence Properties of the EM AlgorithmThe Annals of Statistics, 1983
- A Logistic Model for Paired Comparisons with Ordered Categorical DataBiometrika, 1977