Research Article
Stock Price Prediction Based on Natural Language Processing1
Table 7
Predictive variables after correlation coefficient screening.
| Data type | Words | Lag | Variable | Corr. Coef. |
| Seed keywords | CSI 300 | 1 | | 0.9979 | Inflation rate | 1 | | 0.6903 | Chinese news | 1 | | ā0.6836 | Policy | 10 | | 0.6456 | Dark horse | 10 | | 0.6238 | Stock quotes | 1 | | 0.6130 |
| Generated keywords | CSI 300 | 1 | | 0.9979 | Compound interest | 1 | | 0.7296 | Hot money | 1 | | 0.7096 | Dividend | 1 | | 0.6703 | Profit | 1 | | 0.6513 | Annual interest | 2 | | 0.6218 |
|
|