Research Article
A Risk-Free Discount Rate Prediction Model for Mineral Project Evaluation Using a Hybrid Discrete Wavelet Transform and Artificial Neural Network
Table 6
Optimal training and testing R and MSE results GRNN.
| Frequency component | Width parameter | Training | Testing | R | RMSE | R | RMSE |
| a4 | 0.10 | 0.9972 | 1.876 × 10−1 | 0.9803 | 1.882 × 10−1 | d1 | 0.10 | 0.7938 | 7.122 × 10−2 | 0.6982 | 2.663 × 10−2 | d2 | 0.10 | 0.8675 | 7.678 × 10−2 | 0.8012 | 1.656 × 10−2 | d3 | 0.11 | 0.9428 | 8.909 × 10−2 | 0.8795 | 5.157 × 10−2 | d4 | 0.10 | 0.9810 | 5.024 × 10−2 | 0.9496 | 2.616 × 10−2 |
|
|