Application of Simple Exponential Smoothing Techniques for Forecasting Birth Rates in Maiduguri, Borno State

Abstract

The Specialist Hospital of Maiduguri Metropolitan Council, one of the most popular hospitals in the state where births occur variously delivers women and their newborn babies, the choice of the hospital to conduct this research. The research covers a period of ten years, from 2013 to 2022. The aim is to determine a functional model suitable for forecasting future baby births. This will assist the government in planning for infants, nursing mothers, and projections for infrastructure and human resource development. Different smoothing constants using alpha (α) to find the best model for forecasting baby births. Specifically, smoothing constants of α = 0.1, 0.2, and 0.5 were used in the analysis. The results indicate that a higher smoothing constant provided the best model for the forecasted births. A smoothing constant of α = 0.5 appears to provide the most accurate forecast, as its figures are closer to the observed values than the other smoothing constants. Efforts were also made to determine the smoothing constant that minimizes the error. Towards this end, MAE, MAPE, and MSE were calculated to find the smoothing constant that minimizes the error. In the process, it was noticed that α = 0.9 gives the lowest values of Mean Absolute Error (MAE) and Mean Absolute Percentage Error (MAPE), and therefore was selected as the best smoothing constant that minimizes the error.