Grey adaptive-network-based fuzzy inference system for fund volatility forecasting

A grey adaptive-network-based fuzzy inference system (G-ANFIS) is proposed which combines the grey forecasting model (GM (1, 1)) with the ANFIS and is applied to forecasting fund market volatility in China. A range-based measure of ex-post volatility is employed as a proxy for the unobservable volatility process. The empirical results show that for the RMSE, MAE, LL, LINEX and Mincer-Zarnowitz regression test, the GANFIS approach outperforms the ANFIS and the GM (1, 1) model, which indicates that the G-ANFIS approach is expected to be important in developing the novel strategies for volatility trading and advanced risk management.