Research Article
A Comparative Study of Activation Functions of NAR and NARX Neural Network for Long-Term Wind Speed Forecasting in Malaysia
Table 4
Supportive outcome of other studies.
| Studies | ANN Model | Objective | Accuracy with tansig | Accuracy with logsig |
| M. Rezaeianzadeh et al. [25] | Multi-Layer Perceptron (MLP) | To predict daily outflow | ( = 0.89 and RMSE = 1.69) | ( = 0.80 and RMSE = 2.30) |
| M. Vafaeipour et al. [24] | MLP | To predict Wind velocity | (MAE = 1.48, RMSE = 1.22 and = 0.843) | (MAE = 1.48, RMSE = 1.218 and = 0.844) |
| M. Rezaeianzadeh et al. [22] | MLP | To forecast daily outflow | ( = 0.87 and RMSE = 1.87) | ( = 0.84 and RMSE = 2.1) |
| R. Muazu Musa et al. [23] | MLP | To identify potential archers of psychological coping skill variables | 94% efficiency | 84% efficiency |
| Aladag, and Hakan [26] | ANN | To forecast the number of outpatient visits | (RMSE = 203.06) | (RMSE = 243.28) |
| GSS Gomes [27] | ANN | To forecast financial time series | (MAPE=20%) | (MAPE =25.7 %) |
|
|