PARAMETRIC IDENTIFICATION OF PARTIALLY OBSERVABLE PROCESSES UNDER CONDITIONS OF LIMITED SCOPE OF DATA SELECTION
The paper contains the general survey of procedure of estimation of a partially observable technological process model’s parameters, which does not rely upon asymptotic properties of statistical estimates. A range of influence of the process components which are not measured and a probability distribution of the measurement errors are assumed to be known.
Keywords: posterior distribution, statistical estimate, linear transformation, quadratic form, multidimensional space
