In this video, see how the forms are mathematically equivalent.
- [Instructor] Let's take a look at the relationship…between the error correction form of the smoothing equation…and it's smoothing form.…I'll start with the error correction form.…That's shown in the first line of the screen.…Y hat sub t plus one is the forecast for the next period,…identified as period t plus one.…To make that forecast, we add the forecast for time t,…that is y hat sub t to the product of alpha,…the smoothing constant, and the error in the prior forecast…symbolized by the Greek letter epsilon sub t.…
The next step is to expand the reference to the error…in the forecast for time t.…We do that by replacing epsilon sub t in the first line,…with y sub t minus y hat sub t in the parenthesis…in the second line.…Y sub t is just the actual observation made at time time t…and y hat sub t is the forecast made for time t.…The difference between the two is the forecast error,…or epsilon sub t.…
Next we distribute the alpha across the actual observation…and the forecast for time t.…That's easily done by simply multiplying out…
- Demonstrate how to evaluate a baseline using a correlogram.
- Identify the drawbacks of using Microsoft Excel’s exponential smoothing tool.
- Explain the different ways you can initialize the first forecast.
- Compare the average raw deviation forecast with the mean absolute deviation forecast method.
- Break down the reasons to use R instead of Excel for exponential smoothing.