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

Unlock the full course today

Join today to access over 22,700 courses taught by industry experts or purchase this course individually.

The trend constant compared to the level constant

The trend constant compared to the level constant

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

Start my 1-month free trial

The trend constant compared to the level constant

“

- [Instructor] The smoothing equation has two equivalent forms, the smoothing form and the error correction form. I say that the two forms are equivalent because given identical baselines, the two forms return identical forecasts. You can use either form and the only reason I'm laboring the issue is if the two forms give you two different points of view on what's actually going on in the process of exponential smoothing. Here are the two forms as applied to simple exponential smoothing, typically used with a baseline that shows neither trend nor seasonality. Here's the smoothing form for the forecast level, l hat sub t equals alpha times l sub t minus one plus one minus alpha times l hat sub t minus one. That is the new forecast at time t is the sum of alpha times the observation at time t minus one plus one minus alpha times the forecast at time t minus one. Now we restrict the value of alpha to the range zero to one. Alpha is always a fraction between zero and one and therefore one…

Contents