Statistical Methods for Predicting Malaria Incidences Using Data from Sudan

Table 5

Models, parameter estimates, and fit statistics for malaria incidence: North Kordofan State.

Model	Variable	Coefficient	Std. error	-statistic	Prob.		AIC	MAE
ARIMA		841.29	26.04	32.31	<0.001	0.75	13.13	427.14
	AR(1)	1.72	0.01	128.39	<0.001
	AR(2)	−0.99	0.01	−73.14	<0.001
	MA(1)	−1.70	0.01	−166.85	<0.001
	MA(2)	0.98		81859.26	<0.001
Exponential smoothing		62.05	31.98	1.94	0.059	0.45	13.20	211.28
	AR(1)	0.64	0.18	3.47	0.001
	AR(2)	−0.95	0.23	−4.12	<0.001
	MA(1)	−0.72	0.384	−1.87	0.068
	MA(2)	1.54	0.40	3.83	<0.001
Transformation		0.01		195.40	<0.001	0.73	−13.61	0.0003
	AR(1)	1.73	0.01	126.89	<0.001
	AR(2)	−1.00	0.02	−65.39	<0.001
	MA(1)	−1.70	0.03	−58.41	<0.001
	MA(2)	0.98	0.00	334.60	<0.001
Moving average		−3.03	0.62	−4.85	<0.001	0.25	9.55	47.95
	AR(1)	0.44	0.14	3.11	0.003
	MA(1)	−0.97	0.03	−34.56	<0.001