Statistical Methods for Predicting Malaria Incidences Using Data from Sudan

Table 6

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

Model	Variable	Coefficient	Std. error	-statistic	Prob.		AIC	MAE
ARIMA		404.53	23.69	17.08	<0.001	0.75	11.34	661.05
	AR(1)	1.92	0.16	11.80	<0.001
	AR(2)	−1.28	0.29	−4.36	<0.001
	AR(3)	0.31	0.16	1.92	0.063
	MA(1)	−0.99	0.12	−7.95	<0.001
	MA(2)	−0.93	0.03	−28.42	<0.001
	MA(3)	0.93	0.12	7.64	<0.001
Exponential smoothing		443.49	18.95	23.40	<0.001	0.87	10.60	50.76
	AR(1)	1.73	0.06	29.87	<0.001
	AR(2)	−0.99	0.066	−16.42	<0.001
	MA(1)	−1.09	0.15	−7.35	<0.001
	MA(2)	−0.69	0.105	−6.67	<0.001
	MA(3)	1.55	0.17	9.27	<0.001
Transformation		0.01	0.00	75.27	<0.001	0.66	−12.67	0.0003
	AR(1)	0.69	0.14	4.76	<0.001
	AR(2)	−0.49	0.15	−3.27	0.002
	MA(1)	0.25	0.08	3.075	0.004
	MA(2)	0.41	0.06	6.96	<0.001
	MA(3)	0.86	0.08	10.71	<0.001
Moving average	AR(1)	−0.89	0.22	−4.08	<0.001	0.23	9.85	21.90
	AR(2)	−0.49	0.14	−3.56	0.001
	MA(1)	0.65	0.23	2.86	0.007