Nonlinear extrapolation algorithm for realization of a scalar random process (Q1407341)

The author proposes an algorithm whose purpose is to increase the accuracy of extrapolation of realization of a stochastic process taking into account nonlinear random dependencies that exist in an investigated process and are described by mixed central moment functions. As an example, the problem of nonlinear extrapolation is solved for a moment function of third order. The method is based on Pugachev canonical decomposition apparatus. Comparison results for error estimates of nonlinear and linear extrapolation procedures are presented and advantages of the first one are demonstrated.