Minimaksowo-odporna estymacja parametrów regresji

• Autorzy: Andrzej Grzybowski

Warianty tytułu

Minimax-Robust Estimation of Regression Parameters

• Języki publikacji: PL

Abstrakty

EN

The problem of minimax-robust linear regression estimation is considered under the assumption that prior knowledge about the regression parameter is available for the decision-maker. The prior knowledge is represented by a given class of probability distributions of the unknown parameter. It is also assumed that a covariance matrix of observations is unknown but belongs to a given set of matrices. Some general results are obtained fot the problem under arbitrary quadratic loss. (original abstract)

Słowa kluczowe

PL

Regresja liniowa; Estymacja; Optymalizacja decyzji

EN

Linear regression; Estimation; Decision optimization

• Czasopismo: Przegląd Statystyczny
• Rocznik: 1997
• Tom: 44
• Numer: z. 3
• Strony: 427--435

Twórcy

autor

Andrzej Grzybowski

• Politechnika Częstochowska

Bibliografia

• [1] Bartosiewicz J., Wykłady ze Statystyki Matematycznej, PWN, Warszawa 1989.
• [2] Berger J., A robust generalized bayes estimator and confidence region for a multivariate normal mean, Ann. Statist 8 (1980), s. 716-761.
• [3] Berger J., Selecting a minimax estimator of a multivariete normal mean, Ann. Statist. 10 (1982), s. 81-92.
• [4] Berger J., Robust Bayesian analysis: sensitivity to the prior, J. Statist. Plann. Inference 25 (1990), s. 303-328.
• [5] Berger J., Chen S.Y., Minimaxity of empirical Bayes estimators derived from subjective hyperpriors. Advances in Multivariete Statistical Analysis, 1-12, Ed. A.K. Gupta, Riedel Publ. Comp. (1987), s. 1-12.
• [6] Ferguson T.S., Mathematical Statistics - a decision theoretic approach, Academic Press, New York, 1967.
• [7] Grzybowski A., Minimax control of a system with actuation errors, Applicationes Mathematicae 21, 1 (1991), s. 235-252.
• [8] Grzybowski A., Minimax state estimation for stochastic systems with an uncertain parameter, Applicationes Mathematicae 21, 2 (1991), s. 33-42.
• [9] Läuter H., A minimax linear estimator for linear parameters under restrictions inform of inequalities, Math. Oper. Stat. 6 (1975), s. 689-695.
• [10] Moreno E., Cano J.A., Robust Bayesian analysis with e-contaminations partialy known, J. Roy. Statist. Soc. Ser. B 53-1 (1991), s. 143-155.
• [11] Pilz J., Minimax linear regression estimation with symetrie parameter restictions, J. Statist. Plann. Inference 13 (1986), s. 297-318.
• [12] Pinelis I.F., O minimaksnom ocenivanii regressii, Teorija Verojatnostej i ee primenenija, 35(3) (1991), s. 494-505.
• [13] Rao C.R., Modele liniowe statystyki matematycznej, PWN, Warszawa 1982.
• [14] Verdu S., Poor H.V., On minimax robustness: A general approach and applications, IEEE Trans. Inform. Theory IT-30 (1984), s. 328-340.