From the course: Business Analytics: Forecasting with Trended Baseline Smoothing

Unlock the full course today

Join today to access over 22,400 courses taught by industry experts or purchase this course individually.

Using ARIMA(0,1,1) for SES

Using ARIMA(0,1,1) for SES

From the course: Business Analytics: Forecasting with Trended Baseline Smoothing

Start my 1-month free trial

Using ARIMA(0,1,1) for SES

- [Instructor] ARIMA is a forecasting technique that's closely related to exponential smoothing. ARIMA is an acronym for Auto-Regressive Integrated Moving Average. It's important to get a sense of how some ARIMA models are equivalent to some smoothing models because the ARIMA identification phase can help you decide which smoothing model to use. In particular this chapter has discussed converting a trended baseline to a stationary or horizontal baseline by means of differencing. Then we put the stationary baseline through simple exponential smoothing to more accurately forecast the differences. And we add the forecasted differences back in. In this way we can get forecasts of a trend in the baseline from simple exponential smoothing. We can get the equivalent results from an ARIMA analysis. We'll need to work in R. We need the object named BaseDiffs, which you will have if you saved the workspace at the end of O one O four. If you didn't save it, you can always open the O one O four…

Contents