An Empirical Net Asset Value Forecasting Model based on Optimized ANN using Elephant Herding Strategy

Abstract

Net asset value (NAV) prediction of mutual funds is one of the promising tasks of financial time series data forecasting. It enables the investors to choose the desired mutual fund for investing. Artificial neural network (ANN) is well suited for NAV prediction as the NAV data are nonlinear in nature. This paper proposes the ANN model hybridised with elephant herding optimisation (EHO) algorithm to predict the NAV of different interval days ahead for two of the Indian mutual funds. The prediction performance of ANN-EHO model is compared with ANN, ANN-GA, ANN-PSO and ANN-DE. The results implicate that ANN-EHO model is superior to other four models.