Application of Fuzzy Logic Systems and Fuzzy Neural Networks in Forecasting Problems in Macroeconomy and Finance

Abstract

This chapter is devoted to numerous applications of fuzzy neural networks in economy and financial sphere. In the Sect. 4.2 the problem of forecasting macroeconomic indicators of Ukraine with application of FNN is considered. The goal of this investigation was to estimate the efficiency of different fuzzy inference algorithms. Fuzzy algorithms of Mamdani, Tsukamoto and Sugeno were compared in forecasting Consumer Price Index (CPI) and GDP of Ukraine. As result of this investigation the best forecasting algorithm is detected. In the Sect. 4.3 the problem of forecasting in the financial markets with application of FNN is considered. The forecasting variable are share prices of “Lukoil” at the stock exchange RTS. In these experiments the same FNN are used as in the Sect. 4.2 and the results of these experiments confirmed the conclusions of Sect. 4.2. In the Sect. 4.4 the problem of forecasting corporations bankruptcy risk under uncertainty is considered and investigated. At first classical method by prof. Altman is presented, its properties and drawbacks are analyzed. As an alternative matrix method based on fuzzy sets theory is described and discussed. The application of fuzzy networks with inference algorithms of Mamdani and Tsukamoto for corporations bankruptcy risk forecasting is considered and the experimental comparative results of application all the considered methods for solution of this problem for Ukrainian enterprises are presented and analyzed which confirmed the preference of FNN. In the Sect. 4.5 the problem of banks financial state analysis and bankruptcy forecasting under uncertainty is considered. For its solution the application of FNN TSKBankruptcy risk forecasting FNN TSK and ANFIS is suggested. The experimental results of the efficiency of considered FNN for forecasting are presented and analyzed and the best class of FNN with the least MSE and MAPE for Ukrainian banks bankruptcy forecasting is determined. In the Sect. 4.6 the similar problem of banks financial state analysis and bankruptcy forecasting for leading European banks is considered. The application of FNN for its solution is investigated and the experimental results are presented and analyzed.