The Hybrid GMDH-Neo-fuzzy Neural Network in Forecasting Problems in Financial Sphere

Abstract

The hybrid evolving GMDH-neo-fuzzy system was suggested and investigated. The application of GMDH based on self-organization principle enables to build optimal structure of neo-fuzzy system and train weights of neural network in one procedure. The suggested approach allows to prevent the drawbacks of deep learning such as vanishing or exploding of gradient. As a node of neo-fuzzy system neo-fuzzy neuron with small number of tunable parameters is suggested. This enables to cut training time and accelerate convergence of training. The experimental studies of hybrid neo-fuzzy network were carried out in the task of forecasting of industrial output index, share prices and NASDAQ index. The forecasting efficiency of the suggested hybrid neo-fuzzy system in macro-economy and financial sphere was estimated and its sensitivity to variation of tuning parameters was investigated.