Based On Deep Learning Stock Forecasting And Application

With the rapid development of national economy,the improvement of people's living standard,the status of capital market is constantly improving,and the state is constantly on the road of financial system reform.Finance can not only provide low-cost funds to the real enterprises,but also help them develop rapidly.It can also let the people participate in the stock market and share the dividend of China's economic development.The most direct form of participation is the stock market.Therefore,exploring the law of the stock market operation can not only make the majority of investors profit from it,but also grasp the economic situation,providing some valuable reference for the state when formulating policies.In recent years,due to the rapid development of artificial neural network in the analysis and prediction of the stock market,there are a lot of data processing methods,especially in the stock industry,which has a large amount of data,and has a high application space.Based on the deep learning theory,this paper combines a new model to predict the trend of the stock market by using the characteristics of the deep learning model in different situations when processing data,and then explores the feasibility of the prediction model in the environment of quantitative trading.The main work of this paper is from the following two aspects:On the one hand,the price information of all walks of life always affects the trend of market economy.Stock forecasting is one of the research topics of market economy.Only by forecasting the trend of the stock market,the dynamic information of all walks of life will be presented.However,the information about the market economy contains a lot of noise and uncertainty,which makes stock market forecasting a challenging task.Integrated learning and deep learning are the main methods to solve this problem.Due to the progress of computer technology,there has been a great improvement in the processing of data vacancy and data quality.This paper proposes a method based on the combination of two deep learning models to predict the stock price changes.In this method,convolutional neural network(CNN)and Extreme Gradient Boosting(XGboost)are combined to build a new combination model.The results show that the two methods have better performance in domestic and foreign stock markets.On the other hand,it explains the development potential and practical application of quantitative trading in the stock market from the aspect of algorithm,and makes a mathematical explanation for the application of assets in quantitative trading.The purpose is to verify the theory and practice mutually,so as to achieve a system of decision-making and operation.The orientation of this study is to start with a series of stock combinations,use the long-term and short-term memory model(LSTM)to predict the trend of the stock,and use the reinforcement learning algorithm to judge the stock trading,so as to realize the collaborative trading system of algorithm prediction and quantitative trading.Finally,the whole process is analyzed with the optimization platform and the experimental results are obtained.