Application of filtering equations to problems of statistics of random sequences
The objective of this chapter is to show how the equations of optimal nonlinear filtering obtained for conditionally Gaussian random sequences can be applied to solving various problems of mathematical statistics. In particular, the present section deals with the problem of linear estimation of unobservable components of a multidimensional stationary wide-sense process (discrete time) with rational spectral density in the components accessible for observation.
KeywordsStationary Sequence Recursive Equation Gaussian Vector Recursive Computation Gaussian Sequence
Unable to display preview. Download preview PDF.
Notes and references
- Laning G. H., Battin R. G., Random Processes in Automatic Control Problems. Russian transi., IL., Moscow, 1958.Google Scholar
- Krassovsky N. N., Lidsky A. A., Analytical design of regulators in systems with random properties. III. Optimal regulation in linear systems—The minimal meansquare error. Avtomatika i telemekhanika 11 (1961), 1425–1431.Google Scholar
- Letov A. M., Analytical design of regulators. I–IV. Avtomatika i telemekhanika 4 (1960), 436–441 ;Google Scholar
- Aoki M., Optimization of Stochastic Systems. Russian transl., “Nauka,” Moscow, 1971.Google Scholar
- Wonham W. M., Stochastic Problems in Optimal Control. Tech. report 63–14, Research Institute for Advanced Study, Baltimore, 1963.Google Scholar
- Zhykovsky Ye. L., Liptser R. S., On recursive computing schemes of normal solutions of linear algebraic equations. YVM i MF 12, 4 (1972), 843–857.Google Scholar