Abstract
Statistical decision theory is used to consider the problem of estimating (predicting) a random multidimensional matrix based on the measurement of another random multidimensional matrix in the case of quadratic loss function. An expression for the minimum average risk for the optimal polynomial multidimensional-matrix predictor is derived. The properties of such a predictor are proved and the concept of its efficiency is introduced.
Similar content being viewed by others
References
V. S. Mukha and K. S. Korchyts, “Statistical weather forecasting,” in: Computer Data Analysis and Modeling. Robustness and Computer Intensive Methods, Proc. of the 7th Intern. Conf., Minsk, September 6–10, 2004, 2 (2004), pp. 142–145.
V. S. Mukha, “Statistical vector prediction of the quantitative characteristics of weather,” in: Information Systems and Technologies (IST’2004), Proc. Intern. Conf., Minsk, Belarus, 8–10 November, 2004, Pt. 2 (2004), pp. 195–200.
V. S. Mukha, “The best polynomial multidimensional-matrix regression,” Cybern Syst. Analysis, 43, No. 3, 427–432 (2007).
V. S. Mukha, Statistical Methods of Data Processing: A Handbook [in Russian], Izd. Tsentr BGU, Minsk (2009).
V. S. Mukha, Analysis of Multidimensional Data [in Russian], Tekhnoprint, Minsk (2004).
N. P. Sokolov, Introduction to the Theory of Multidimensional Matrices [in Russian], Naukova Dumka, Kyiv (1971).
Author information
Authors and Affiliations
Corresponding author
Additional information
Translated from Kibernetika i Sistemnyi Analiz, No. 2, pp. 121–130, March–April 2011.
Rights and permissions
About this article
Cite this article
Mukha, V.S. Minimum average risk and efficiency of optimal polynomial multidimensional-matrix predictors. Cybern Syst Anal 47, 277–285 (2011). https://doi.org/10.1007/s10559-011-9309-3
Received:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s10559-011-9309-3