Predictability Of Shanghai Composite Index Based On Neural Network And Sentiment Index

With the economic development and transformation of people’s sense aboutinvesting, the stock market has become an important part of modern life. The stockmarket investment has become one of the hot public issues. The quality of the stockmarket has become the focused point of managers and investors. The benefit andrisk of the stock market investment is often directly proportional to higherinvestment income, the greater the risk of possible risk. Stock market predictionmethod has an extremely important value and theoretical significance, but thecomplexity of the internal structure of the price system, the variability of externalfactors determine the enormity of the task, the traditional forecasting tools have notmet this need.In this paper, on the basis of behavioral finance theory and neural networkmodeling theory research, select trading prices and investor sentiment that reflectthese two dimensions of market information and price formation process to establisha BP neural network prediction model of to test the effect of the application of theBP neural network to predict the Shanghai Composite Index. Researchers usuallyselect a technical indicator on investor sentiment as trader sentiment proxy variab les.In this paper, a new sentiment indicators, principal component analysis method toconstruct a new trader sentiment index as a predictive model input variables.Meanwhile, use the linear model to fit the Shanghai Composite Index, as acomparative model to detect the predicted effect of the neural network.The theoretical analysis and experimental results prove that the BP neuralnetwork prediction is feasible for the Shanghai Composite Index. The sentimentindex added significantly improve the precision of the model. The sentiment indexhas some value in the field of stock index forecast.