Stock market risk prediction intelligent implementation method based on deep learning

Abstract

The invention discloses a stock market risk prediction intelligent implementation method based on deep learning. According to the method, more abstract high-feature representation is formed by combining shallow-level features, so that a deep-level implicit relationship of data can be discovered, and a composite function can be better obtained by stacking multiple layers of neural networks and selecting a Sigmoid activation function. According to the method, self-learning self-adaption is carried out through a long-term and short-term memory (LSTM) network model in deep learning, and accordingto training of past historical data, accurate prediction of the whole stock market in the future can be well obtained. According to the method, more abstract high-feature representation is formed by combining shallow-level features so as to discover a deep-level implicit relationship of data, and a composite function can be better obtained by stacking multiple layers of neural networks and selecting a reasonable activation function. Good technical effects are generated on multiple indexes such as the RMSE, the error value and the self-designed profit value.

Description

technical field [0001] The invention relates to an intelligent realization method of stock market risk prediction based on deep learning, which belongs to the field of communication technology. Background technique [0002] The stock market plays a very important role in the financial market, and in the stock market, whether it can accurately predict the future fluctuations of the overall stock market is also the key to obtaining income. In recent years, investors have tried to use relevant methods such as expert analysis, combination analysis, and strategy research to predict stocks. Most of them use the relevant content of probability and statistics, and use a variety of mathematical models at the same time, and finally get a prediction result. However, the trend of stock prices is affected by many factors such as national policies, national economic conditions, international environment and business operating conditions, so a single mathematical model cannot accurately d...

AI Technical Summary

Problems solved by technology

However, SVM technology is difficult to implement for large-scale training samples, and the data in the stock market is large and diverse. When the amount of data becomes larger, the training time of SVM will become longer, which will bring greater consumption to the prediction. time cost

In addition, the classic support vector machine algorithm only gives the algorithm of the two-class classification method, but in the practical application of stock data mining, it is generally necessary to solve multi-class problems

In addition, for SVM technology, it is sensitive to missing data, sensitive to the selection of parameters and kernel parameters, and the selection of these parameters is generally artificial, with a certain degree of randomness and non-representation. In the field of stock market, it should be Domain knowledge is quoted, but there is no good way to solve the problem of kernel function selection

Images