Abstract
We consider the linear regression model where prior information in the form of linear inequalities restricts the parameter space to a polyhedron. Since the linear minimax estimator has, in general, to be determined numerically, it was proposed to minimize an upper bound of the maximum risk instead. The resulting so-called quasiminimax estimator can be easily calculated in closed form. Unfortunately, both minimax estimators may violate the prior information. Therefore, we consider projection estimators which are obtained by projecting the estimate in an optional second step. The performance of these estimators is investigated in a Monte Carlo study together with several least squares estimators, including the inequality restricted least squares estimator. It turns out that both the projected and the unprojected quasiminimax estimators have the best average performance.
Similar content being viewed by others
References
Judge G. G. and Takayama T.: 1966, Inequality restrictions in regression analysis,J. Amer. Statist. Assoc. 61, 166–181.
Judge G. G. and Yancey T. A.: 1986,Improved Methods of Inference in Econometrics, North-Holland, Amsterdam.
Lemke C. E.: 1965, Bimatrix equilibrium points and mathematical programming,Management Sci. 11, 681–689.
Leskinen, E.: 1981, Simulation notes on Farebrother's minimum mean squared error estimator, Reports on Statistics 2/1981, Department of Statistics, University of Jyväskylä.
Pilz J.: 1986, Minimax linear regression estimation with symmetric parameter restrictions,J. Statist. Plann. Inference 13, 297–318.
Ravindran A.: 1972, A computer routine for quadratic and linear programming problems,Comm. ACM 15, 818–820.
Schmidt K.: 1992,Parameterschätzung im linearen Regressionsmodell bei Vorinformation in Ungleichungsform (Parameter Estimation in Linear Regression under Inequality Restrictions), Mathematical Systems in Economics, Vol. 128, Hain, Frankurt.
Schmidt K. and Stahlecker P.: 1989, Zur Ausnutzung von a priori Informationen bei der Schätzung einer gesamtwirtschaftlichen Cobb-Douglas-Produktionsfunktion (On the estimation of a macroeconomic Cobb-Douglas production function using prior information),Jahrbücher für Nationalökonomie und Statistik 206, 599–605.
Schmidt K. and Stahlecker P.: 1995, Reducing the maximum risk of regression estimators by polyhedral projection,J. Statist. Comput. Simulation 52, 1–15.
Stahlecker P., Jänner M. and Schmidt K.: 1991, Linear-affine Minimax-Schätzer unter Ungleichungsrestriktionen (Linear-affine minimax estimators under inequality restrictions),Allgemeines Statistisches Archiv 75, 245–264.
Stahlecker, P. and Trenkler, G.: 1985, On heterogeneous versions of the best linear and the ridge estimator, in T. Pukkila and S. Puntanen (eds),Proc. First Tampere Seminar on Linear Models, pp. 301–322.
Stahlecker P. and Trenkler G.: 1993, Minimax estimation in linear regression with convex polyhedral constraints,J. Statist. Plann. Inference 36, 185–196.
Trenkler G. and Stahlecker P.: 1987, Quasi minimax estimation in the linear regression model,Statistics 18, 219–226.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Schmidt, K. A comparison of minimax and least squares estimators in linear regression with polyhedral prior information. Acta Appl Math 43, 127–138 (1996). https://doi.org/10.1007/BF00046993
Received:
Issue Date:
DOI: https://doi.org/10.1007/BF00046993