UDC 519.2
DOI: 10.36871/2618-9976.2021.02.004
Authors
Zvyagin Leonid Sergeevich
PhD of Economics, Associate Professor, Associate Professor of the Department of «System Analysis
in Economics», Financial University under the Government of the Russian Federation, Moscow, Russia
Abstract
In the research, modeling, and forecasting of complex systems associated with various natural processes, scientists have encountered uncertainty and unpredictability. In order to eliminate or reduce the impact of uncertainty on the prediction of complex natural processes and to determine possible catastrophic consequences, traditional methods and models are not applicable and do not show reliable results, since they do not take into account the stochastic component. The article deals with the use of Bayesian theory in various spheres of life, and will also consider and highlight the factors that have a significant impact on the activation and flow of the processes and phenomena under consideration. The Bayes trust network is constructed and trained by filling in tables of conditional probabilities by experts and introducing numerical values of factors to determine the unconditional (a priori) probabilities of the factors under consideration. The constructed network can be supplemented, if necessary, with vertices and connections between them.
Keywords
Probabilistic estimates
Bayesian approach
Decisionmaking
Mathematical methods
Modeling
