An ARMA Model for Short-term Prediction of Hepatitis B Virus Seropositivity among Blood Donors in Lafia-Nigeria

In this paper, we attempt to search for an optimal Autoregressive Moving Average (ARMA) model that best forecast hepatitis B virus infection among blood donors in Lafia-Nigeria. The study uses monthly data in Lafia-Nigeria for the period of 11 years 6 months from January 2007 to June 2018. The data was obtained as secondary data from General Hospital Lafia and Dalhatu Araf Specialist Hospital, Lafia. The time series and stationarity properties of the data are explored using time plots and Dickey-Fuller Generalized Least Squares unit root test. The results indicate that the series is integrated of order zero, I(0). An ARMA (p,q) model in line with Box-Jenkins procedure was employed to model the time series data. The result shows that ARMA (1,1) was the best candidate to model and forecast hepatitis B virus infection among blood donors in Lafia- Nigeria. Critical analysis of the model shows that the HBV infection is chronic among blood donors in the study area. The estimated ARMA (1,1) model was then used to forecast future values of hepatitis B infection among blood donors in Lafia-Nigeria from July 2018 to June 2019. The forecast shows a stable level of infection for the forecasted period. The study provided some policy recommendations.