Forecasting of Unemployment and Economic Growth for Turkey: ARIMA Model Application

Abstract

The main purpose of the study is to obtain the forecasting values of unemployment rate and economic growth in the coming years. Since unemployment and economic growth series are not stationary series in level I(0), the preferred model for forecasting is the Autoregressive Integrated Moving Average (ARIMA) model. Models determined for unemployment and economic growth forecasting with the help of model measurement criteria are ARIMA (2,1,1) and ARIMA (1,1,0) models respectively. In the study, the period between 1988-2017 has been considered as the prediction period and the forecasting values obtained for this period have been compared graphically with the actual values and the success of the forecasting has been evaluated. Since the forecasting power of the models is successful, forecasts have been made for the 2018-2019 (ex-post) period and it has been determined that the error rate between forecasting values and actual values is at a level to be considered good. Forecasting values for the 2020-2025 (ex-ante) period have been designed, it has been observed that unemployment rates will increase in a fluctuating manner in the coming years and economic growth is in a constant rising trend.