On admissibility in estimating the mean squared error of a linear estimator

• Autorzy: Stępniak, Czesław
• Języki publikacji: EN

Abstrakty

EN

Consider a linear estimator of a parametric vector Cß in the normal Gauss-Markov model Y ~ N {Xß, σV). The Mean Squared Error of such an estimator may be presented in the form kσ+ß' X' KXß and may be estimated by quadratics in Y Some basic decision-theoretic questions in the estimation are discussed. Among others, some admissible estimators of the MSE in several classes of the quadratics are given.

Słowa kluczowe

EN

linear estimator; estimators of MSE; Gauss-Markov model

• Czasopismo: Probability and Mathematical Statistics
• Rocznik: 1998
• Tom: Vol. 18, Fasc. 1
• Strony: 33--38
• Opis fizyczny: Bibliogr. 12 poz.

Twórcy

autor

Stępniak, Czesław

• Department of Mathematics, Agricultural University of Lublin, P.O. Box 158, PL-20-950 Lublin, Poland

Bibliografia

• [1] J. О. Berger and C. Robert, Subjective hierarchical Bayes estimation of a multivariate normal mean, Ann. Statist. 18 (1990), pp. 617-651.
• [2] J. R. Brook and T. Moore, On the expected length of the least squares coefficient vector, J. Econometrics 12 (1980), pp. 245-246.
• [3] S. Gnot, G. Trenkler and R. Zmyślony, Nonnegative minimum biased quadratic estimation in the linear regression models, J. Multivariate Anal. 54 (1995), pp. 113-125.
• [4] M. Grządziel, A simulation study of a new variable selection strategy based on minimum biased quadratic estimation, Discussiones Mathematicae. Algebra and Stochastic Methods 15 (1995), pp. 325-333.
• [5] Dr A. Harville, Alternative formulations and procedures for the two-way mixed model, Biometrics 34 (1978), pp, 441-453.
• [6] J. H. Klotz, R. C. Milton and S. Zacks, Mean square efficiency of estimators of variance components, J. Amer. Statist. Assoc. 64 (1969), pp. 1383-1402.
• [7] L. R. LaMotte, On non-negative quadratic unbiased estimation of variance components, ibidem 68 (1973), pp. 728-730.
• [8] — Some results on biased linear estimation applied to variance components estimation, in: Mathematical Statistics and Probability: Proceedings. Sixth International Conference, Wisła, Poland, Springer, New York 1980, pp. 266-274.
• [9] M. D. Perlman, Reduced mean square error estimation for several parameters, Sankhyä В 34 (1972), pp. 89-92.
• [10] A. L. Rukhin, Quadratic estimators of quadratic functions of normal parameters, J. Statist. Plann. Inference 15 (1987), pp. 301-310.
• [11] J. Seely, A complete sufficient statistic for the linear model under normality and singular covariance matrix, Comm. Statist. Theory Methods A 7 (1978), pp. 1465-1473.
• [12] C. Stępniak, On admissibility in various classes of quadratic estimators, Statist. Probab. Lett. 4 (1986), pp. 17-19.