Forecasting Analysis of Unemployment Rate in Malaysia based on the Naive, Holt's Linear Trend and Box-Jenkins Models

Abstract

Unemployment has become one of the most vital challenges to the economy in most of the developed and developing countries, along with the socio-economic problem. Generally, the unemployment rate is the key element to measure whether a country is doing a good job of achieving productive employment or not. The previous studies mainly focused on forecasting quarterly and yearly unemployment rates by using Simple Exponential Smoothing, Holt's Linear Trend and ARIMA model. Yet, there are not many studies that focus on forecasting the monthly unemployment rate in Malaysia. Consequently, this study aims to compute the best model among the Naive model, Holt's Linear trend model and the Box-Jenkins model for forecasting the monthly unemployment rate in Malaysia. The results revealed that the Naive model was the best model with the lowest error rates of Mean Square Error (MSE), Mean Absolute Deviation (MAD) and Mean Absolute Percentage Error (MAPE). The forecast value for 3 months ahead unemployment rate was found to be 3.3%.