Research Article
A Risk-Free Discount Rate Prediction Model for Mineral Project Evaluation Using a Hybrid Discrete Wavelet Transform and Artificial Neural Network
Table 5
Optimal training and testing R and MSE results RBFNN.
| | Frequency component | Width parameter | Number of neurons | Training | Testing | | R | RMSE | R | RMSE |
| | a4 | 1.2 | 18 | 1.0000 | 4.02 × 10−3 | 1.000 | 3.97 × 10−3 | | d1 | 0.5 | 6 | 0.7469 | 7.66 × 10−2 | 0.7404 | 2.40 × 10−2 | | d2 | 0.3 | 14 | 0.8384 | 8.24 × 10−2 | 0.8176 | 1.49 × 10−2 | | d3 | 0.4 | 9 | 0.9547 | 7.42 × 10−2 | 0.9698 | 2.57 × 10−2 | | d4 | 0.9 | 20 | 0.9948 | 2.54 × 10−2 | 0.9947 | 8.43 × 10−3 |
|
|
