A new adaptive network-based fuzzy inference system with adaptive adjustment rules for stock market volatility forecasting

Forecasting stock volatility is crucial to many fundamental problems of financial field, such as risk management, asset pricing and asset allocation etc. This paper proposes a new Adaptive Network-Based Fuzzy Inference System (ANFIS) which adaptively adjusts fuzzy inference rules by using Fruit Fly Optimization Algorithm (FOA). Empirical analysis is made on the Shanghai A-share sample stocks. Compared with ANFIS, the experimental results reveal that this new model can accurately and successfully forecast the sample stocks' volatility.