| Summary: | The paper considers two directions of development of the dynamic-stochastic approach to the construction and use of predictive models. The first direction is related to the uncertainty of the initial state of the simulated process, and the second-to the stochastic nature of model parameter estimates. In the first case, we consider methods for calculating fast-growing perturbations (FGPs) of the initial state of atmospheric dynamics models and a method for using FGPs in optimizing observation systems based on information ordering. We describe all the main details of the mathematical apparatus that allow us to use these methods in predicting any multidimensional processes and in optimizing any spatial monitoring systems. An example of determining the zones of the Northern hemisphere where errors in measurements of the initial state of the atmosphere most significantly affect the accuracy of the forecast is given. In the second case, a mathematical apparatus for generating perturbations of model parameters in accordance with their probability distribution is proposed. Based on the data of the USSR economic indices, a numerical example of estimation and perturbation of the parameters of the Volterra model is given. The analysis of the results of integration of the original and modified models is given. The paper contains significant additions to the published earlier author's papers, at the same time being a logical continuation of these papers.
|
|---|
