Risk modeling extremum methods multiobjective optimization

Abstract

Purpose: the article is devoted to the theoretical, methodological and instrumental support for mathematical modeling, analysis and prediction of risk of extreme values in the natural and economic systems. Discussion: developed methods of obtaining information before forecast based on classical statistics methods such as mathematical statistics, multicriteria optimization, extreme value theory. Effectiveness of the proposed approach is demonstrated by specific time series of volumes of mountain rivers. Results: the work is devoted to the study of multi-criteria optimization methods and classical statistics before forecast obtain information for time series that have long-term memory, which is why their levels do not meet the independence property. The complexity of modeling extreme values necessitates the use of new methods, namely the methods of nonlinear dynamics.