Probability Density Forecasting Method Based on Neural Network Quantile Regression and Kernel density estimation

• 1、School of Management，HeFei University of Technology
• 2、School of Management，HeFei University of Technology
• 3、Hui Merchants Bank Lu\'an Branch,Lu’an China

Abstract：This paper introduced the neural network quantile regression and kernel density estimation method toforecast stock price in the future, which combined the powerful nonlinear adaptive ability of neural network and the advantage of quantile regression descripting explanatory variables ; and then，using the kernel density estimation method obtains probability density curve of the future stock price. Meanwhile, the application of probability forecasting density curve gets the corresponding stock price at the highest probability point value, which can obtain higher prediction accuracy. We can not only obtain the day's stock price prediction probability density value, can also obtain the interval of the future value of the stock price changes. The method can provide more quantitative basis and value of information to investors when they make investment decisions with the help for investors to create greater value.

Keywords： Stock price prediction quantile regression neural network kernel density estimation probability density