Time series stock price prediction using recurrent error based neuro-fuzzy system with momentum

Abstract

Stock market analysis is very important not only for making profit or averting big losses, but also to recognize the direction of the market. The direction point of the market has significant effects on capital investment, other business cycle issues and socio-economical level of the country. This study proposes a new approach for stock market price prediction using recurrent error based neuro-fuzzy system with momentum (RENFSM). The experiment found that the proposed model can provide superior performance for stock market price prediction than ANFIS and traditional recurrent type ANFIS networks.