HYBRID MODELS FOR FORECASTING INDICATORS OF FOREIGN ECONOMIC ACTIVITY OF THE RUSSIAN FEDERATION BASED ON FUZZY COGNITIVE MAP AND DECISION TREES

UDC 336.018 (045)
DOI: 10.36871/2618-9976.2023.01.001

Authors

Sergei A. Yarushev,
Candidate of Technical Sciences, Associate Professor, Plekhanov Russian University of Economics, Moscow, Russia
Alexey N. Averkin,
Candidate of Physical and Mathematical Sciences, Associate Professor, Leading Researcher, FRC IU RAS, Plekhanov Russian University of Economics, Moscow, Russia
Victoria M. Savinova,
Senior Lecturer, Department of Informatics, Plekhanov Russian University of Economics, Moscow, Russia
Vladislav R. Iksanov,
Postgraduate student of the Department of Informatics, Plekhanov Russian University of Economics, Moscow, Russia
Mikhail A. Buduev,
Student, Department of Informatics, Plekhanov Russian University of Economics, Moscow, Russia

Abstract

The development of hybrid forecasting models is more relevant than ever during periods of economic crisis, since it is thanks to hybrid technologies that it is possible to make forecasts even in unstable times. In this paper, several indicators of the foreign economic activity of the Russian Federation were selected for experiments, in total, 5 indicators were selected, the data on which were presented quarterly for the period from 2013 to 2021. For forecasting, two cognitive forecasting models were used, such as decision trees and fuzzy cognitive models. cards.

Keywords

Forecasting, Regression decision tree, Fuzzy cognitive maps, Macroeconomics, Artificial intelligence, Cognitive modeling