Predicting Research Net Value Of Fund Based On Neural Network

With the development of society and economy,fund has become one of themost important investment tools. Security investment funds, an advanced systemarrangement and attractive financial tool,has drawn widely attention from investorsaround the world. The performance of security investment funds not only reflectsfiduciary duty, but also offers useful decision-making information. It is a manual forfund managers and a guide for fund investors. It is very important for thedevelopment of fund industry and the whole security market. As an investment toolwith high risk and high return, the risk of fund is mainly considered as the fluctuationin price. Fund price reflects the over all fluctuation condition of the prices in the fundmarket, a correct prediction on fund price is helpful to make decision on fundinvestment.A back-propagation neural network is applied to forecast the net asset value(NAV) of JINTAI by identification characteristic of neural network. The results of theforecasting show that the neural network model has good nonlinear reflection abilityand learning ability. It is effective and applicable in forecasting NAV tendency andinflexion of funds.The theoretical significance lies in that it explores a new risk analysis andassessment technique of fund investment. It provides a quantitative analysistechnique of fund risk analysis and assessment based on multi-factor system. It offersa scientifically comprehensive method based on BP Neural Network. Finally itpresents individual investors and institution investors a new train of thought whenthey make certain investment decisions.