For the same time series you used in the previous exercise, try using a nonseasonal model applied to the seasonally adjusted data obtained from STL. The stlf() function will make the calculations easy...

For the same time series you used in the previous exercise, try using a nonseasonal model applied to the seasonally adjusted data obtained from STL. The stlf() function will make the calculations easy (with method="arima" ). Compare the forecasts with those obtained in the previous exercise. Which do you think is the best approach?