Abstract
The optimality of linear estimates in minimax estimation of a stochastically uncertain vector in a linear observation model by mean-square criterion is studied. In the Gaussian case, a uniformly optimal linear estimate is shown to exist in the class of all unbiased estimates. Moreover, it is minimax in the class of all nonlinear estimates if the nonrandom parameters of the observation model are unbounded. If the a priori information on random parameters are given as constraints on the covariance matrix, linear estimates are shown to be minimax.
Similar content being viewed by others
REFERENCES
Marchuk, A.G. and Osipenko, K.Yu., Best Approximation of Functions Defined with Error at a Finite Number of Points, Mat. Zametki, 1975, vol. 17, no. 3, pp. 359.
Magaril-Il'yaev, G.G. and Osipenko, K.Yu., Optimal Estimation of Functionals from Inexact Data, Mat. Zametki, 1991, vol. 50, no. 6, pp. 85–93.
Kurzhanskii, A.B., Upravlenie i otsevivanie v usloviyakh neopredelennosti (Control and Estimation under Uncertainty), Moscow: Nauka, 1977.
Gusev, M.I., Optimality of Linear Algorithms in Guaranteed Estimation, Izv. Ross. Akad. Nauk, Tekh. Kibern., 1994, no. 3, pp. 87–95.
Matasov, A.I., Optimality of Linear Estimation Algorithms of Guaranteed Estimation: I, Kosm. Issl., 1988, vol. 26, no. 5, pp. 643–653.
Matasov, A.I., Optimality of Linear Estimation Algorithms of Guaranteed Estimation: II, Kosm. Issl., 1988, vol. 26, no. 6, pp. 807–812.
Liptser, R.Sh. and Shiryaev, A.N., Statistika sluchainykh protsessov, Moscow: Nauka, 1974. Translated under the title Statistics of Random Processes, Berlin: Springer-Verlag, 1978.
Vandelinde, V.D., Robust Properties of Solutions to Linear-Quadratic Estimation and Control Problems, IEEE Trans. Autom. Control, 1977, vol. 22, no. 1, pp. 138–139.
Verdú, S. and Poor, H.V., Minimax Linear Observers and Regulators for Stochastic Systems with Uncertain Second-Order Statistics, IEEE Trans. Autom. Control, 1984, vol. 29, no. 6, pp. 499–511.
Verdú, S. and Poor, H.V., On Minimax Robustness: A General Approach and Applications, IEEE Trans. Inf. Theory, 1984, vol. 30, no. 2, pp. 328–340.
Pankov, A.R., Recurrent Conditionally Minimax Filtration of Processes in Nonlinear Stochastic Difference Systems, Izv. Ross. Akad. Nauk, Tekh. Kibern., 1992, no. 3, pp. 71–77.
Solov'ev, V.N., Minimax Bayes Estimation Theory, Teor. Veroyatn. Primenen., 1999, vol. 44, no. 4, pp. 738–756.
Pankov, A.R. and Semenikhin, K.V., Methods of Parametric Identification of Multidimensional Linear Models under a priori Uncertainty, Avtom. Telemekh., 2000, no. 5, pp. 76–92.
Pankov, A.R., Minimax Methods of Estimation and Optimization of Processes in Stochastically Uncertain Systems, Doctoral Dissertation, Moscow: Mosk. Aviat. Inst., 1998.
Zacks, Sh., The Theory of Statistical Inference, New York: Wiley, 1971. Translated under the title Teoriya statisticheskikh vyvodov, Moscow: Mir, 1975.
Borovkov, A.A., Matematicheskaya statistika (Mathematical Statistics), Moscow: Nauka, 1984.
Ibragimov, I.A. and Khas'minskii, R.Z., Asimptoticheskaya teoriya otsenivaniya (Asymptotic Estimation Theory), Moscow: Nauka, 1977.
Demidenko, E.Z., Lineinaya i nelineinaya regressii (Linear and Nonlinear Regressions), Moscow: Finansy i Statistika, 1981.
Solov'ev, V.N., Optimality of Linear Algorithms in Guaranteed Estimation under Random Measurement Errors, Kosm. Issl., 1994, vol. 32. no. 2, pp. 122–124.
Matasov, A.I., Optimality of Linear Algorithms in the "Worst Correlation" Problem, Vestn. Mosk. Gos. Univ., 1989, no. 1, pp. 61–63.
Albert, A., Regression, and the Moor-Penrose Pseudoinverse, New York: Academic, 1972. Translated under the title Regressiya, psevdoinversiya i rekurrentnoe otsenivanie, Moscow: Nauka, 1977.
Pyt'ev, Yu.P., Metody analiza i interpretatsii dannykh (Data Analysis and Interpretation Methods), Moscow: Mosk. Gos. Univ., 1990.
Kuks, A. and Ol'man, V., A Minimax Linear Estimate for Regression Cofficients, Izv. Akad. Nauk ESSR, Fiz. Mat., 1972, vol. 21, no. 1, pp. 66–72.
Author information
Authors and Affiliations
Rights and permissions
About this article
Cite this article
Semenikhin, K.V. The Optimality of Linear Estimation Algorithms in Minimax Identification. Automation and Remote Control 65, 493–503 (2004). https://doi.org/10.1023/B:AURC.0000019382.18600.73
Issue Date:
DOI: https://doi.org/10.1023/B:AURC.0000019382.18600.73