Application of Regularized GRU-LSTM Model in Stock Price Prediction

The stock market is a highly complex nonlinear movement system, and its fluctuation law is affected by many factors, so the prediction of the stock price index is a very challenging task. There are many examples showing that Neural Network algorithms are well suited for such time series predictions and often achieve satisfactory results. In this paper, based on the existing models, we proposed a Regularized GRULSTM neural network model and applied it to the short-term forecast of closing price of the two stocks. The experimental results show that our proposed model is superior to the existing GRU and LSTM network models in stock time series prediction.