Analysis and Modeling of Prevalence of Measles in the Ashanti Region of Ghana
Kwame Asare Gyasi-Agyei
Department of Applied Mathematics, Koforidua Polytechnic, Koforidua, Ghana.
William Obeng Denteh *
Department of Mathematics, Kwame Nkrumah University of Science and Technology, Kumasi, Ghana.
Amoakoh Gyasi-Agyei
Ghana ICT Research Institute (GICTRAC), PO Box AH163, Accra, Ghana.
*Author to whom correspondence should be addressed.
Abstract
In this paper, autoregressive integrated moving average (ARIMA) model is used to predict the prevalence and incidence of measles in the Ashanti Region of Ghana. The Mean Absolute Error (MAE) and the Mean Squared Error (MSE) are used to compare the in-sample forecasting performance of four selected competing models. The working data from the Ashanti Health Services spans from 2001 to 2011. It is evident from the analysis that measles data in the Ashanti Region of Ghana could best be modeled with ARIMA (2, 1, 1) and that measles prevalence in the Ashanti Region is expected to increase if no preventative measures are taken. The forecasting accuracy using MAE for ARIMA (2, 1, 1) is calculated as 28.1141 and the forecasting accuracy using MSE for ARIMA (2, 1, 1) is calculated as 2947.15.
Keywords: Measles, prevalence, models, forecasting, autoregressive, virus, vaccine, symptoms