Comparative study of smoothing methods and box-jenkins model in forecasting unemployment rate in Malaysia / Norliana Mohd Lip … [et al.]

One of the significant problems faced by all countries in the world, especially developing countries is unemployment. It must be taken into account by the government regardless of whether its effect is critical or under control. According to the Department of Statistics Malaysia (DOSM), the unemployment rate refers to the proportion of the unemployed population to the total population in labour force. The high unemployment rate will reflect the negative impact on labour market. This study was conducted to identify the best model among smoothing techniques (double exponential smoothing and Holt’s model) and the Box-Jenkins model for forecasting the unemployment rate in Malaysia since there is no appropriate model in estimating the unemployment rate in the future. The performance of the best model was determined by using several error measures such as Mean Square Error (MSE), Root Mean Square Error (RMSE), Mean Absolute Percentage Error (MAPE) and Geometric Root Mean Square Error (GRMSE). The lowest error measure identified which model is appropriate to forecast the unemployment rate in Malaysia. The findings showed that the most appropriate model is ARIMA (2,1,3) in the Box-Jenkins method since it indicated the smallest value of all error measures. In addition, the study demonstrated that there was an increment in the unemployment rate in Malaysia since the forecasted trend line showed low fluctuation which gradually increasing and decreasing from January 2019 until December 2019. This implied that the future value of the unemployment rate in Malaysia is stable and not considered as a critical problem.