Deep-learning-based method for predicting stock market

Abstract

The invention provides a deep-learning-based method for predicting a stock market. The method comprises the following steps of data pretreatment and construction of an LSTM-DCC stock market predictionmodel. According to the invention, on the basis of combination of advantages of an expansion cause-and-effect convolution network and a recurrent neural network, data features of an own stock sequence and correlated influence factors are learned automatically and thus the future price and trend are predicted.

Description

technical field [0001] The patent of this invention relates to a stock market price prediction method based on deep learning. Background technique [0002] The financial field is the center of gravity of a country's economic operation, and the stock market can also reflect the overall economic situation of a country in one aspect, and with the development of the country's economy, people pay more attention to capital management, and investment and financial management have become a norm. Therefore, it is of great practical significance to analyze and model the time series data of the stock market and predict the stock market trend in the future. [0003] Compared with general time series data (temperature, wind, etc.), stock market time series data has the characteristics of high data dimensionality, nonlinearity, and non-stationary characteristics. These characteristics make the prediction of time series data for the stock market a difficult and hot topic in research. . ...

AI Technical Summary

Problems solved by technology

[0012] Traditional stock market forecasting methods are usually difficult to effectively extract good data features, making the accuracy of forecasting results often low.

Images