The Unifying Role of Iterative Generalized Least Squares in Statistical Algorithms
作者： Guido del Pino
关键词:
摘要: This expository paper deals with the role of iterative generalized least squares as an algorithm for computation statistical estimators. Relationships between various algorithms, such Newton-Raphson, Gauss-Newton, and scoring, are studied. A parallel is made properties model structure numerical employed to find parameter estimates. In particular a general linearizability property that extends concept link function in linear models considered its computational meaning discussed. Maximum quasilikelihood estimators reinterpreted so they may exist even when there no function.
jstor.org 
sci-hub.se 参考文章(28)
R. Thompson, R. J. Baker, Composite Link Functions in Generalized Linear Models Journal of The Royal Statistical Society Series C-applied Statistics. ,vol. 30, pp. 125- 131 ,(1981) , 10.2307/2346381
Daryl Pregibon, Score Tests in GLIM with Applications Springer, New York, NY. pp. 87- 97 ,(1982) , 10.1007/978-1-4612-5771-4_9
Peter McCullagh, John Ashworth Nelder, Generalized Linear Models ,(1983)
Avriel, M, Nonlinear Programming: Analysis and Methods ,(1976)
Birch B. Jeffrey, Some convergence properties of iterated reweighted least squares in the location model Communications in Statistics - Simulation and Computation. ,vol. 9, pp. 359- 369 ,(1980) , 10.1080/03610918008812162
D. F. Andrews, A Robust Method for Multiple Linear Regression Technometrics. ,vol. 16, pp. 523- 531 ,(1974) , 10.1080/00401706.1974.10489233
Peter J. Huber, Robust statistical procedures ,(1977)
Robert I. Jennrich, Asymptotic Properties of Non-Linear Least Squares Estimators Annals of Mathematical Statistics. ,vol. 40, pp. 633- 643 ,(1969) , 10.1214/AOMS/1177697731
R. W. M. WEDDERBURN, Quasi-likelihood functions, generalized linear models, and the Gauss-Newton method Biometrika. ,vol. 61, pp. 439- 447 ,(1974) , 10.1093/BIOMET/61.3.439
Murray Aitkin, A Note on the Regression Analysis of Censored Data Technometrics. ,vol. 23, pp. 161- 163 ,(1981) , 10.1080/00401706.1981.10486259