Skip Navigation

Biometrika 1995 82(2):426-432; doi:10.1093/biomet/82.2.426
© 1995 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 CRIBARI-NETO, F.
Right arrow Articles by FERRARI, S. L. P.
Right arrow Search for Related Content
Social Bookmarking
Add to CiteULike   Add to Connotea   Add to Del.icio.us  
What's this?

MISCELLANEA

Second order asymptotics for score tests in generalised linear models

FRANCISCO CRIBARI-NETO and SILVIA L. P. FERRARI

Department of Economics, Southern Illinois University Carbondale, Illinois 62901-4515, U.S.A.
Departamento de Estatistica, Universidade de São Paulo Caixa Postal 20570, São Paulo/SP, 01452-990, Brazil

This paper develops finite-sample corrections for score tests in generalised linear models when the dispersion parameter is unknown, thus generalising the results of Cordeiro, Ferrari & Paula (1993). We show that the coefficients which define the Edgeworth expansion for the null distribution of the score statistic when the dispersion parameter is unknown are the coefficients obtained by these authors plus some extra terms. We also give closed-form expressions for such coefficients. An important special case of our results is the normal linear regression model. Simulation results for this model show that the corrections can be very effective, reducing the probability of conflict with other asymptotically equivalent testing criteria.

Key Words: Asymptotic expansion • Bartlett-type correction • Chi-squared distribution • Generalised linear model • Score test • Size-correction


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




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.