Abstract
We consider the linear regression modely=Xβ+u with prior information on the unknown parameter vector β. The additional information on β is given by a fuzzy set. Using the mean squared error criterion we derive linear estimators that optimally combine the data with the fuzzy prior information. Our approach generalizes the classical minimax procedure firstly proposed by Kuks and Olman.
Similar content being viewed by others
References
Bandemer, H., Gottwald, S. (1993): Einführung in Fuzzy-Methoden (4th ed.). Berlin (Akademie Verlag).
Christopeit, N., Helmes, K. (1996): Linear Minimax Estimation With Ellipsoidal Constraints. Acta Applicandae Mathematicae (P. Stahlecker, G. Trenkler, Eds.). Dordrecht (Kluwer Academic Publishers), 3–15.
Drygas, H. (1982): Minimax prediction in linear models. Linear Statistical Inference (W. Klonecki, T. Calinski, Eds.). New York (Springer Verlag), 48–60.
Drygas, H. (1993): Reparametrization Methods in Linear Minimax Estimation. Stat. Sci. & Data Anal., 87–95.
Drygas, H., Pilz, J. (1996): On the Equivalence of Spectral Theory and Bayesian Analysis in Minimax Linear Estimation. Acta Applicandae Mathematicae (P. Stahlecker, G. Trenkler, Eds.). Dordrecht (Kluwer Academic Publishers), 43–57.
Gaffke, N., Heiligers, B. (1989): Bayes, Admissible, and Linear Minimax Estimators in Linear Models with Restricted Parameter Space. Statistics20, 487–508.
Girko, V.L. (1996): Spectral Theory of Minimax Estimation. Acta Applicandae Mathematicae (P. Stahlecker, G., Trenkler, Eds.). Dordrecht (Kluwer Academic Publishers), 59–69.
Heiligers, B. (1993): Linear Bayes and minimax estimation in linear models with partially restricted parameter space. Journal of Statistical Planning and Inference36, 175–184.
Hoffmann, K. (1979): Characterization of minimax linear estimators in linear regression. Mathematische Operationsforschung und Statistik, Series Statistics10, 19–26.
Kuks, J., Olman, W. (1972): Minimax linear estimation of regression coefficients II (in Russian). Izv. Akad. Nauk. Eston. SSR21, 66–72.
Lauterbach, J., Stahlecker, P. (1992): A numerical method for an approximate minimax estimator in linear regression. Linear Algebra and its Applications176, 91–108.
Läuter, H. (1975): A minimax linear estimator for linear parameters under restrictions in form of inequalities. Mathematische Operationsforschung und Statistik, Series Statistics6, 689–695.
Pilz, J. (1986): Minimax linear regression estimation with symmetric parameter restriction. Journal of Statistical Planning and Inference13, 297–318.
Pilz, J. (1991): Bayesian estimation and experimental design in linear regression models. New York (Wiley).
Stahlecker, P. (1987): A priori Information und Minimax Schätzung im Linearen Regressionsmodell (in German). Mathematical Systems in Econom. 108. Frankfurt (Athenäum).
Stahlecker, P., Jänner, M., Schmidt, K. (1991): Linearaffine Minimax-Schätzer unter Ungleichungsrestriktionen (in German). Allgemeines Statistisches Archiv75, 245–264.
Stahlecker, P., Trenkler, G. (1991): Linear and Ellipsoidal Restrictions in Linear Regression. Statistics22, 163–176.
Teräsvirta, T. (1989): Estimating linear models with incomplete ellipsoidal restrictions. Statistics20, 187–194.
Toutenburg, H. (1982): Prior Information in Linear Models. New York (Wiley).
Toutenburg, H., Roeder, B. (1978): Minimax-linear and Theil estimator for restrained regression coefficients. Mathematische Operationsforschung und Statistik, Series Statistics9, 409–505.
Trenkler, G., Stahlecker, P. (1987): Quasi minimax estimation in the linear regression model. Statistics18, 219–226.
Zadeh, L.A. (1965): Fuzzy Sets. Information and Control8, 338–353.
Zimmermann, H.-J. (1991): Fuzzy Set Theory-and Its Applications (2nd ed.). Dordrecht (Kluwer Academic Publishers).
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Arnold, B.F., Stahlecker, P. Fuzzy prior information and minimax estimation in the linear regression model. Statistical Papers 38, 377–391 (1997). https://doi.org/10.1007/BF02925995
Received:
Revised:
Issue Date:
DOI: https://doi.org/10.1007/BF02925995