Abstract
With the development of recent years, the field of deep learning has made great progress. Compared with the traditional machine learning algorithm, deep learning can better find the rules in the data and achieve better fitting effect. In this paper, we propose a hybrid stock forecasting model based on Feature Selection, Convolutional Neural Network and Bidirectional Gated Recurrent Unit (FS-CNN-BGRU). Feature Selection (FS) can select the data with better performance for the results as the input data after data normalization. Convolutional Neural Network (CNN) is responsible for feature extraction. It can extract the local features of the data, pay attention to more local information, and reduce the amount of calculation. The Bidirectional Gated Recurrent Unit (BGRU) can process the data with time series, so that it can have better performance for the data with time series attributes. In the experiment, we used single CNN, LSTM and GRU models and mixed models CNN-LSTM, CNN-GRU and FS-CNN-BGRU (the model used in this manuscript). The results show that the performance of the hybrid model (FS-CNN-BGRU) is better than other single models, which has a certain reference value.
Highlights
With the development of China’s economy and the improvement of people’s living standards, the stock market has become a hot area of attention
The results show that the model has good universality and stability, and can provide a certain reference value for many investors and research institutes. [3] proposed a feature selection algorithm based on weighted sum least squares support vector machine (LS-SVM)
The results show that the neural network can deal with continuous and classified forecasting variables. [10] adopts a hybrid model (RNN + LSTM)and the results show that the hybrid model has a good application prospect for the stock price forecast of single stock with variables such as corporate behavior and corporate announcement. [11] proposed a new appearance model, which can be embedded into the recurrent neural network of bidirectional short-term memory unit and can effectively learn to track
Summary
With the development of China’s economy and the improvement of people’s living standards, the stock market has become a hot area of attention. Scholars and investors in various countries are very optimistic about the future development trend of China’s stock market. The change of stock price is influenced by many factors, such as natural disasters, the influence of politicians and the influence of the country in the world. Because the change of stock price is nonlinear, it is very important for many researchers and investors [1] to predict the trend of stock price in advance. The establishment of a high precision and reasonable stock prediction model can effectively reduce the loss of investors in the stock market, and can improve their control of the stock price.
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have