Share Price Trend Prediction Using CRNN with LSTM Structure
碩士 === 國立勤益科技大學 === 資訊工程系 === 106 === The stock market plays a very important role in the entire financial market, and one of the most attractive research issues in predicting stock price fluctuations. In this thesis, we will use the historical information of stocks to predict future stock prices an...
| Main Authors: | , |
|---|---|
| Other Authors: | |
| Format: | Others |
| Language: | zh-TW |
| Published: |
2018
|
| Online Access: | http://ndltd.ncl.edu.tw/handle/bjerqu |
| Summary: | 碩士 === 國立勤益科技大學 === 資訊工程系 === 106 === The stock market plays a very important role in the entire financial market, and one of the most attractive research issues in predicting stock price fluctuations. In this thesis, we will use the historical information of stocks to predict future stock prices and use deep learning to achieve this.
This paper uses deep learning to predict the trend of future stock prices. Since the trend of stocks is usually related to the previous stock price, this paper proposes the architecture of ConvLSTM based on the convolutional recurrent neural network (CRNN), which uses the architecture of long short-term memory (LSTM) in RNN. LSTM improves the long-term dependence of traditional RNNs and effectively improves the accuracy and stability of stability.
This paper collected a total of ten stock historical data to test, and achieved an average error rate of 3.449 RMSE.
|
|---|