Relying on Holt’s linear exponential smoothing model forecasts to inform the national HIV policy in Eswatini

Abstract

This study uses annual time series data of HIV prevalence among individuals aged 15-49 years for Eswatini from 1990 to 2020 to predict future trends of HIV prevalence over the period 2021 to 2030. The study utilizes Holt’s linear exponential smoothing model. The optimal values of smoothing constants α and β are 0.9 and 0.4 respectively based on minimum MSE. The results of the study indicate that annual HIV prevalence among individuals aged 15-49 years will continue to decline over the out of sample period. Therefore, we encourage authorities to address major drivers of HIV transmission among young adults with more focus being given to HIV prevention interventions