A GENERALIZED QUASI-LIKELIHOOD METHOD FOR ESTIMATION OF AUTOREGRESSIVE MODELS

by A. A. Abed Al-Aziz, S .M. El- Sayed and Y. M. Selim.

Abstract: In this paper, generalized quasi-likelihood estimation fora semi-parametric model with a dependent structure is developed incorporating knowledge of skewness and kurtosis. This is a generalization of a similar idea proposed for independent observations by Godambe and Thompson (1989). The generalized quasi-score function is constructed on the basis of non-orthogonal estimating function invoking the method of Durairajan (1992).

Key Words: Generalized Linear Models; Quasi- likelihood method; Non-orthogonal estimating functions

Authors:
A. A. Abed Al-Aziz, Alaa_mnn@yahoo.com
S .M. El- Sayed, smelsayed@yahoo.com
Y. M. Selim, yousefselim@yahoo.com

Editor: Bradley T Ewing, bewing@ba.ttu.edu

READING THE ARTICLE: You can read the article in portable document (.pdf) format (99771 bytes.)

NOTE: The content of this article is the intellectual property of the authors, who retains all rights to future publication.

This page has been accessed 3072 times since July 24, 2006.


Return to the InterStat Home Page.