From the course: Business Analytics: Forecasting with Exponential Smoothing
Join today to access over 22,700 courses taught by industry experts or purchase this course individually.
The least squares approach
From the course: Business Analytics: Forecasting with Exponential Smoothing
The least squares approach
- [Instructor] The mean absolute deviation is one way to avoid positive and negative forecasts canceling one another out. Recall that it converts each forecast error to its absolute value. A different approach is to simply square each forecast error. The square of any number is positive, therefore, if you squared the errors, you'll wind up with a sequence of positive numbers that will not cancel one another out, and that's what you see on this worksheet. One difference is simple terminology. It's traditional to term the difference between a forecast and its associated observation as a deviation in the context of mean absolute deviations. It's also traditional to refer to that difference as error; in the context that we're working with here, the root mean squared error. Columns A through E are again identical to columns A through E in the prior lesson. In column F, however, instead of taking the absolute value of a deviation or error, we square it. Cell G3 is the average of those…
Practice while you learn with exercise files
Download the files the instructor uses to teach the course. Follow along and learn by watching, listening and practicing.
Download courses and learn on the go
Watch courses on your mobile device without an internet connection. Download courses using your iOS or Android LinkedIn Learning app.