Stock market movement prediction using LDA-online learning model
© 2018 IEEE. In this paper, an online learning method namely LDA-Online algorithm is proposed to predict the stock movement. The feature set which are the opening price, the closing price, the highest price and the lowest price are applied to fit the Linear Discriminant Analysis (LDA). Experiments o...
Saved in:
| Main Authors: | , |
|---|---|
| 其他作者: | |
| 格式: | Conference or Workshop Item |
| 出版: |
2019
|
| 主題: | |
| 在線閱讀: | https://repository.li.mahidol.ac.th/handle/123456789/45600 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 機構: | Mahidol University |
| 總結: | © 2018 IEEE. In this paper, an online learning method namely LDA-Online algorithm is proposed to predict the stock movement. The feature set which are the opening price, the closing price, the highest price and the lowest price are applied to fit the Linear Discriminant Analysis (LDA). Experiments on the four well known NASDAQ stocks (APPLE, FACBOOK GOOGLE, and AMAZON) show that our model provide the best performance in stock prediction. We compare LDA-online to ANN, KNN and Decision Tree in both Batch and Online learning scheme. We found that LDA-Online provided the best performance. The highest performances measured on GOOGLE, AMAZON, APPLE FACEBOOK stocks are 97.81%, 97.64%, 95.58% and 95.18% respectively. |
|---|