Stock prediction method introducing investor preferences

Abstract

The invention relates to the technical field of stock investment, in particular to a stock prediction method for introducingintroducing investor preferences, which utilizes a foreground theoretical value to describe investor behavior preferences and introduces an index into prediction of stock income. The method comprises the following steps: S1, calculating a foreground theoretical value by utilizing historical yield data of stocks; Ss2, carrying out LSTM model construction for introducing investor preferences is carried out; Ss3, predicting a model process.

Description

technical field [0001] The invention relates to the technical field of stock investment, in particular to a stock prediction method introducing investor preferences. Background technique [0002] Research on stock forecasting methods mainly revolves around statistical methods and machine learning methods. The most commonly used statistical methods include autoregressive conditional heteroscedasticity (ARCH) model, generalized autoregressive conditional heteroscedasticity (GARCH) model, autoregressive moving average (ARIMA) model, etc. However, statistical forecasting methods have certain limitations. In recent years, some scholars have studied stock forecasting from a new perspective in order to improve the accuracy of forecasting. Gong et al. introduced corporate social responsibility to establish a stock forecasting model, and the results showed that corporate social responsibility has a positive effect on stock price increases, and can improve the accuracy of stock fore...

AI Technical Summary

Problems solved by technology

[0006] To sum up, although domestic and foreign scholars have conducted sufficient research on the relationship between investor behavior preferences and stock returns, no scholars have yet used investor behavior preferences as a predictor of stock returns.

Images