Modeling the number of unemployed in South Sumatra Province using the exponential smoothing methods.

Abstract

The number of open unemployment in South Sumatra Province from year to year is found to be unstable. It can cause serious developmental problems. One solution to this problem is to build an early warning system by predicting the number of open unemployment in the future so that the Regional Government can establish relative policies to anticipate the negative impacts it will have on the environment, economy, social and politics. Therefore, this study discusses the best model to predict the number of unemployed in South Sumatra Province. The methods used to identify the best model are Single Exponential Smoothing (SES), Brown’s Exponential Smoothing (BES), and Holt’s Exponential Smoothing (HES). The Exponential Smoothing methods are compared to obtain forecasting results with a minimal error rate. Root Mean Square Error (RMSE) and Mean Absolute Percentage Error (MAPE) metrics are used to measure the performance of the forecasting model. Empirical results show that the SES model with the smoothing parameter value = 0.7 is the best significant model in predicting the number of open unemployment in South Sumatra Province with a MAPE value of 6.24% and an RMSE value of 23.058. Thus, this SES model can be a reference for the Government to predict the number of open unemployment in South Sumatra Province so that the Regional Government can anticipate the negative impacts it can cause.