Bartlett corrections and bias correction for two heteroscedastic regression models
- 1 January 1992
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics - Theory and Methods
- Vol. 22 (1) , 169-188
- https://doi.org/10.1080/03610929308831012
Abstract
We present, in matrix notation, general Bartlett correction formulae for several hypotheses in the multiplicative and dependent variable heteroscedastic regression models. The formulae are simple enough to be used analytically to obtain several closed-form expressions in special cases. They have also advantages for numerical purposes. Improved likelihood ratio tests are discussed for several test problems in these models. By simulation, the performance of the Bartlett corrections is illustrated. We derive general formulae for the n−1 bias of the maximum likelihood estimates in these models, where n is the sample size.Keywords
This publication has 13 references indexed in Scilit:
- On the corrections to the likelihood ratio statisticsBiometrika, 1987
- Diagnostics for heteroscedasticity in regressionBiometrika, 1983
- Robust Tests for Heteroscedasticity Based on Regression QuantilesEconometrica, 1982
- A Test for Heteroscedasticity Based on Ordinary Least Squares ResidualsJournal of the American Statistical Association, 1979
- Using Residuals Robustly I: Tests for Heteroscedasticity, NonlinearityThe Annals of Statistics, 1978
- Correcting Inhomogeneity of Variance with Power Transformation WeightingTechnometrics, 1974
- A New Test for HeteroskedasticityJournal of the American Statistical Association, 1969
- A General Definition of ResidualsJournal of the Royal Statistical Society Series B: Statistical Methodology, 1968
- Some Tests for HomoscedasticityJournal of the American Statistical Association, 1965
- A GENERAL METHOD FOR APPROXIMATING TO THE DISTRIBUTION OF LIKELIHOOD RATIO CRITERIABiometrika, 1956