Modeling Seasonal Influenza Transmission and Its Association with Climate Factors in Thailand Using Time-Series and ARIMAX Analyses
Table 4
Summary of ARIMA model fitting parameters in the central region during 2009–2014.
Model
Fit
Pred.
Climate variables
RMSE
AIC
RMSE
Vars
Coef.
value
() ARIMA(1, 0, 2)(1, 0, 0)12
0.4550
89.74
0.7837
() ARIMAX(1, 0, 2)(1, 0, 0)12 with Rainfall
0.4425
88.01
0.7810
Rainfall (lag 0)
−0.1234
0.046
() ARIMAX(1, 0, 2)(1, 0, 0)12 with Rainfall
0.4420
86.42
0.8339
Rainfall (lag 1)
0.1374
0.0241
() ARIMAX(1, 0, 2)(1, 0, 0)12 with Rainfall
0.4584
89.99
0.8045
Rainfall (lag 2)
0.0643
0.1989
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4344
82.36
0.7042
(lag 3)
−0.2806
0.8523
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4224
84.13
0.8139
(lag 4)
3.9727
0.0356
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4536
91.47
0.7989
(lag 0)
−0.4727
0.6001
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4442
86.95
0.7586
(lag 1)
1.9988
0.0392
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4549
91.73
0.7843
(lag 0)
−0.0469
0.6372
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4170
79.37
0.8880
(lag 1)
2.0353
0.0003
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4439
85.90
0.6760
(lag 2)
−1.3507
0.0256
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4544
91.59
0.7911
(lag 0)
−0.2535
0.7034
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4355
84.52
0.7896
(lag 1)
1.7995
0.0090
() ARIMAX(1, 0, 2)(1, 0, 0)12 with
0.4571
89.21
0.7390
(lag 2)
−0.7104
0.3375
() ARIMAX(1, 0, 2)(1, 0, 0)12 with ,
0.3998
74.98
0.7507
(lag 4) (lag 1)
4.4553 2.4223
0.0001 0.0039
() ARIMAX(1, 0, 2)(1, 0, 0)12 with ,
0.3922
70.78
0.9337
(lag 4) (lag 1)
3.8566 1.8012
0.0267 0.0026
() ARIMAX(1, 0, 2)(1, 0, 0)12 with ,
0.3786
72.82
0.5792
(lag 4) (lag 2)
3.8699 −1.9457
<0.0001 <0.0001
() ARIMAX(1, 0, 2)(1, 0, 0)12 with ,
0.3786
72.14
0.8027
(lag 4) (lag 1)
4.9256 2.0620
<0.0001 <0.0001
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.
