Research Article
Comparison of Conventional Modeling Techniques with the Neural Network Autoregressive Model (NNAR): Application to COVID-19 Data
Table 2
Accuracy of different time-series models for predicting COVID-19 cumulative daily cases.
| | Method | Source | RMSE | MAE |
| | Mean | Training | 1566.384 | 1309.385 | | Testing | 2207.310 | 1862.713 | | Drift | Training | 412.2189 | 284.6854 | | Testing | 2114.5660 | 1713.2369 | | Naïve | Training | 412.2303 | 284.5769 | | Testing | 2177.6843 | 1688.9792 | | Holt | Training | 394.6241 | 277.7294 | | Testing | 2191.3800 | 1740.4823 | | SES | Training | 394.5822 | 277.6888 | | Testing | 2180.9038 | 1744.2653 | | ARIMA order SES | Training | 392.7844 | 275.1783 | | Testing | 2182.1964 | 1754.7370 | | ARIMA (2, 2, 2) | Training | 392.7844 | 275.1783 | | Testing | 2182.1964 | 1754.7370 | | ARIMA auto | Training | 393.0668 | 275.3658 | | Testing | 2182.9865 | 1760.4807 | | NNAR | Training | 195.301 | 143.5501 | | Testing | 2136.069 | 1589.569 |
|
|
