Empirical likelihood for linear regression models under imputation for missing responses
- 18 December 2001
- journal article
- Published by Wiley in Canadian Journal of Statistics
- Vol. 29 (4) , 597-608
Abstract
The authors study the empirical likelihood method for linear regression models. They show that when missing responses are imputed using least squares predictors, the empirical log‐likelihood ratio is asymptotically a weighted sum of chi‐square variables with unknown weights. They obtain an adjusted empirical log‐likelihood ratio which is asymptotically standard chi‐square and hence can be used to construct confidence regions. They also obtain a bootstrap empirical log‐likelihood ratio and use its distribution to approximate that of the empirical log‐likelihood ratio. A simulation study indicates that the proposed methods are comparable in terms of coverage probabilities and average lengths of confidence intervals, and perform better than a normal approximation based method.Keywords
This publication has 9 references indexed in Scilit:
- Empirical Likelihood Confidence Intervals for Linear Regression CoefficientsJournal of Multivariate Analysis, 1994
- Empirical Likelihood and General Estimating EquationsThe Annals of Statistics, 1994
- On the accuracy of empirical likelihood confidence regions for linear regression modelAnnals of the Institute of Statistical Mathematics, 1993
- The Bootstrap and Edgeworth Expansion1992
- Empirical Likelihood for Linear ModelsThe Annals of Statistics, 1991
- Methodology and Algorithms of Empirical LikelihoodInternational Statistical Review, 1990
- Empirical Likelihood Ratio Confidence RegionsThe Annals of Statistics, 1990
- Empirical likelihood ratio confidence intervals for a single functionalBiometrika, 1988
- The Analysis of Categorical Data from Complex Sample Surveys: Chi-Squared Tests for Goodness of Fit and Independence in Two-Way TablesJournal of the American Statistical Association, 1981