MODELING AND FORECASTING NEW DIARRHOEA CASES IN UNDER-FIVES IN CHITUNGWIZA URBAN DISTRICT, ZIMBABWE EMPIRICAL EVIDENCE FROM A BOX-JENKINS CATCH ALL MODEL

Abstract

In this paper, the Artificial Neural Network model has been employed to project diarrhoea cases in children under five years of age in Chitungwiza urban district in Zimbabwe. The study covers the period January 2012 to December 2018. The out-of-sample forecasts range over the period January 2019 to December 2020. The residual analysis of the model indicates that the model applied in this research article is stable and suitable for forecasting under 5 diarrhoea cases in Chitungwiza urban district. The forecasts show a generally downwards trajectory of diarrhoea cases in Chitungwiza urban district over the period January 2019 – December 2020. Amongst other recommendations, there is need for the relevant authorities to strengthen sanitation and hygienic practices in Chitungwiza urban district.