An autoregressive integrated moving average model for short-term prediction of hepatitis C virus seropositivity among male volunteer blood donors in Karachi, Pakistan

Table 1 Hepatitis C virus seroprevalence (1000^-1× year^-1) among male volunteer blood donors at four large blood banks in Karachi (1998- 2005)

Yr	Mean	SD	95% CI for mean		Minimum	Maximum
			Lower limit	Upper limit
1998	13.8	1.5	12.8	14.7	12	16
1999	16.3	2.2	14.8	17.7	13	21
2000	18.5	2.2	17.1	19.9	15	22
2001	19.8	2.0	18.6	21.1	16	22
2002	19.9	3.0	18.0	21.8	15	26
2003	20.3	3.2	18.2	22.3	16	27
2004	29.3	2.6	27.6	30.9	25	33
2005	24.8	2.1	23.4	26.1	21	27
Total	20.3	5.1	19.3	21.3	12	33

F = 47.9; df = 7, 88; P < 0.001.

Table 2 Autoregressive integrated moving average model (2,1,7) of hepatitis C virus seroprevalence (1000^-1× month^-1) among male volunteer donors in Karachi, Pakistan, (January 1998-December 2005)

Parameters	Estimate	Standard error	t-ratio
Autoregressive parameter (Φ)
Φ2	0.67	0.15	4.5
Moving average parameter (θ)
θ2	-0.59	0.18	3.3
θ3	0.49	0.15	3.3
θ4	-0.80	0.11	7.3
θ6	0.74	0.21	3.5
θ7	-0.37	0.17	2.2

White noise variance = 9.74; Ljung-Box Q statistic = 20.4 (P = 0.433).