Stock closing price prediction method

Abstract

The invention discloses a stock closing price prediction method, and relates to the technical field of deep learning and stock prediction. According to the stock closing price prediction method, stock closing price data is input into a constructed stock closing price prediction model, and a prediction result of the stock closing price is output. The stock closing price prediction model comprises a complementary set empirical mode decomposition module, a gating circulation unit module, a full connection layer and width learning which are connected in sequence. The stock closing price prediction method improves the prediction precision of the stock closing price.

Description

technical field [0001] The present invention relates to the technical fields of deep learning and stock forecasting, in particular to a method for forecasting stock closing prices. Background technique [0002] In recent years, the domestic financial industry has developed rapidly, which plays a vital role in promoting the development of the country, and has a huge impact on the stability of the country's industrial and economic development. As the most basic financial tool in the financial system, stocks can most intuitively show the strength of a country's economic development, and the prediction of stock trends has become a major concern of investors. In order to provide investors with meaningful investment decisions and reduce their investment risks, various stock forecasting models have been proposed. It is known that financial data such as stocks have the characteristics of nonlinearity, multi-scale, and multi-noise. Traditional machine learning models have a lot of l...

AI Technical Summary

Problems solved by technology

It is known that financial data such as stocks have the characteristics of nonlinearity, multi-scale, and multi-noise. Traditional machine learning models have a lot of limitations. Large-scale data such as stocks

The proposal of a single deep learning model such as long short-term memory network (LSTM) and gated recurrent unit (GRU) has shown certain advantages in the field of prediction, but there are still problems of low accuracy

In recent years, the time series prediction model combining the recurrent neural network (RNN) and the convolutional neural network (CNN) has been widely used, but the pooling layer of CNN will lose a lot of valuable information during training, ignoring the relationship between the local and the whole sex

Images