Table 5.
Summary of the ARIMA model fitting parameters in southern region during 2009–2014.
| Model | Fit | Pred. | Climate variables | |||
|---|---|---|---|---|---|---|
| RMSE | AIC | RMSE | Vars | Coef. | P value | |
| (1) ARIMA(1, 0, 2)(0, 0, 1)12 | 0.3486 | 57.74 | 0.2496 | |||
| (2) ARIMAX(1, 0, 2)(0, 0, 1)12 with T mean | 0.3415 | 56.65 | 0.3062 | T mean (lag 3) | −1.8054 | 0.6023 |
| (3) ARIMAX(1, 0, 2)(0, 0, 1)12 with T mean | 0.3471 | 56.46 | 0.2235 | T mean (lag 4) | 2.6311 | 0.4453 |
| (4) ARIMAX(1, 0, 2)(0, 0, 1)12 with RHmax | 0.3480 | 58.44 | 0.2748 | RHmax (lag 1) | −1.5168 | 0.6160 |
| (5) ARIMAX(1, 0, 2)(0, 0, 1)12 with T mean, RHmax | 0.3483 | 57.49 | 0.2589 |
T
mean (lag 4) RHmax (lag 1) |
3.3490 3.6432 |
0.3580 0.3640 |
ARIMAX: autoregressive integrated moving average with input series; fit: fitting results; RMSE: root mean square error; AIC: Akaike's Information Criterion; Pred.: prediction of ARIMA model; Coef.: coefficient of climate variables; lag: time lag of climate variables.