Journal article

EXPONENTIAL DISPERSION MODELS AND THE GAUSS‐NEWTON ALGORITHM

GK Smyth

Australian Journal of Statistics | Published : 1991

Abstract

It is known that the Fisher scoring iteration for generalized linear models has the same form as the Gauss‐Newton algorithm for normal regression. This note shows that exponential dispersion models are the most general families to preserve this form for the scoring iteration. Therefore exponential dispersion models are the most general extension of generalized linear models for which the analogy with normal regression is preserved. The multinomial distribution is used as an example. Copyright © 1991, Wiley Blackwell. All rights reserved

University of Melbourne Researchers

Citation metrics