Regression analysis of dependent error models
- 1 October 1986
- journal article
- research article
- Published by Cambridge University Press (CUP) in Bulletin of the Australian Mathematical Society
- Vol. 34 (2) , 199-209
- https://doi.org/10.1017/s0004972700010066
Abstract
A method of analysing the general linear regression model is described, for the case where the observations are correlated. For many applications the correlations are structured, with neighbouring observations being more strongly correlated than those some distance apart in time or space. Such correlation structures may often be assumed to belong to some class of models indexed by a small number of parameters. Estimation and inference procedures which are able to cope with a wide range of correlation models, are described and the methods are applied to problems which occur in biometry.Keywords
This publication has 12 references indexed in Scilit:
- The Levinson Algorithm and Its Applications in Time Series AnalysisInternational Statistical Review, 1984
- Recursive Residuals on a Rectangular LatticeBiometrical Journal, 1984
- GENERALIZED INVERSES USED IN RECURSIVE ESTIMATION OF THE GENERAL LINEAR MODEL1Australian Journal of Statistics, 1983
- Best Linear Recursive Estimation for Mixed Linear ModelsJournal of the American Statistical Association, 1981
- Best Linear Recursive Estimation for Mixed Linear ModelsJournal of the American Statistical Association, 1981
- Estimation in regression models with stationary, dependent errorsCommunications in Statistics - Theory and Methods, 1981
- Stochastic Growth Curve AnalysisPublished by JSTOR ,1979
- Linear Dynamic Recursive Estimation from the Viewpoint of Regresion AnalysisJournal of the American Statistical Association, 1972
- Linear Dynamic Recursive Estimation from the Viewpoint of Regression AnalysisJournal of the American Statistical Association, 1972
- Fractional Brownian Motions, Fractional Noises and ApplicationsSIAM Review, 1968