Forecasting Infant Mortality Rate in Somalia Using Artificial Neural Networks

In this research article, the ANN approach was applied to analyze infant mortality rate in Somalia. The employed data covers the period 1982-2020 and the out-of-sample period ranges over the period 2021-2030. The residuals and forecast evaluation criteria (Error, MSE and MAE) of the applied model indicate that the model is stable in forecasting infant mortality rate in Somalia. The ANN (12, 12, 1) model predicted that infant mortality in the country is likely to surge up to as high as 109/1000 live births per annum around 2027. These results are an early warning of a possible disastrous situation that can be experienced in Somalia if drastic action is not taken now. Therefore the government is encouraged to ensure high coverage of child immunizations, Vitamin A supplementation, exclusive breast feeding of babies for at least 6 months and institutional deliveries. The suggested 7-fold policy directions summarize what the government ought to do in order to address infant mortality in Somalia.