The main purpose of this study is to predict the unemployment rate in Indonesia by
using time series data from 1986 to 2015 using autoregressive integrated moving
average (ARIMA). A differencing process is required due to the actual time series of
the unemployment rates in Indonesia is non-stationary. The results show that the best
model for forecasting the unemployment rate in Indonesia by using the ARIMA (0,2,1)
model. The forecasting results reveal that the unemployment rate in Indonesia tends to
decrease continuously. The average of the residuals is close to zero which informs a
good result of the forecasting analysis.