Stock Forecasting Analysis based on Deep Learning and Quantitative Investment Algorithms with Multiple Indicators

Abstract

Traditional stock forecasting methods are generally based on linear models. However, the price of stocks is affected by a variety of objective factors and does not present a simple linear relationship. Neural network is a good tool for predicting nonlinear data. In order to predict the trend of stock price more accurately, we use neural network prediction method on TensorFlow to consider the nonlinear factors affecting the price of stocks, forecast future data based on past data, and use historical transaction records of stocks to analyze and forecast future prices. In this model, historical data such as technical indicators and fundamental indicators are used as input variables, and deep learning and quantitative investment algorithms with multiple indicators are used to predict the rise and fall trend of stock prices after several days, and build an investment portfolio based on the predicted results. The results show that the correct rate of trend prediction is 83.5%, which has a good prediction effect.

Fuzzy Decision-Making Analysis of Quantitative Stock Selection in VR Industry Based on Random Forest Model

“…Zhang et al [10,11] give empirical tests on the stock price vulnerability using topology network. In addition, more studies on quantitative stock selection models have used classifiers such as GRU neural network models or integrated tree models [12][13][14], and the most comprehensive one is the Stacking method, which combines the abovementioned neural networks, gradient boosting trees, and XGBoost to form a new algorithmic model, RGXB-Stacking stock selection model, and the research results show that this model has significantly better back-testing effect on constituent stock data than other models [15][16][17]. Based on the above analysis of existing research, it is found that most of the research on quantitative stock selection nowadays is only at the level of optimizing the selected impact factors, and there is no good solution for the relationship between a large number of impact factors and the expected return of the model [18].…”