Table 1. Summary of model performance and the estimated coefficients for Hong Kong Influenza.

Model	Fit		Prediction	AR		MA		Environmental variables
	RMSE	AIC	RMSE	Est.	Pr > |t|	Est.	Pr >|t|	Vars	Est.	Pr >|t|
ARIMA(2,1,2)	0.4045	166.26	0.4788	0.44	<.0001	0.588	0.0037
				−0.446	<.0001	−0.785	0.0025
ARIMA(1,1,2)	0.4071	166.25	0.4321	0.45	0.0226	0.603	0.0014
						−0.375	<.0001
SARIMA(1,0,0)(0,1,0)	0.5144	159.46	0.4993	0.774	<.0001
SARIMA(2,0,0)(0,1,0)	0.5074	156.56	0.5033	0.608	<.0001
				0.219	0.0288
ARIMAX(1,1,2) with LST, RF, and RH	0.3675	138.51	0.5292	0.426	0.0181	0.6443	0.0001	LST (Lag2)	−0.035	0.0016
						−0.446	<.0001	LST (Lag5)	−0.0307	0.0049
								RF (Lag 3)	0.0534	0.0047
								RH	0.0164	<.0001
SARIMAX(0,1,2)(1,0,0) with LST, RF and RH	0.3662	137.52	0.5433	0.276	0.0104	0.2937	0.0005	LST (Lag2)	−0.036	0.0011
						−0.316	0.0002	LST (Lag5)	−0.0324	0.0032
								RF (Lag 3)	0.0527	0.0064
								RH	0.0168	<.0001
SARIMAX((2),1,0) with LST	0.4013	156.76	0.4649	0.251	0.0022			LST (Lag2)	−0.028	0.014
								LST (Lag5)	−0.0256	0.0257
SARIMAX(1,0,0)(0,1,0) with LST	0.4666	134.7	0.5104	0.795	<.0001			LST (Lag2)	−0.048	0.0009
								LST (Lag5)	−0.0312	0.0248
ARIMAX(2,1,0) with RF	0.4174	168.62	0.4029	0.244	0.0017			RF (Lag 3)	0.0244	0.1985
ARIMAX((2),1,0) with RH	0.4073	163.34	0.4728	0.247	0.0021			RH (Lag 1)	0.011	0.0055
SARIMAX(1,0,1)(0,1,0) with RH	0.4968	152.76	0.5831	0.883	<.0001	0.2588		RH (Lag1)	0.0144	0.0086
ARIMAX((2),1,0) with LST and RH	0.3872	148.07	0.5273	0.259	0.0018			LST (Lag2)	−0.029	0.0096
								LST (Lag 5)	−0.029	0.0097
								RH (Lag 1)	0.0124	0.0013

Abbreviations: ARIMA  =  Autoregressive Integrated Moving Average; S = Seasonal; X = with external/input series; LST  =  Land Surface Temperature; RF  =  Accumulated Rainfall; RH  =  Relative Humidity; RMSE  =  Root Mean Square Error; AIC  =  Akaike's Information Criterion; AR  =  Autoregressive coefficients; MA  =  Moving Average Coefficients; Est  =  Estimated values through conditional least square method.