Our last class of the course.
Music “Still Alive” by Jonathan Coulton
Here are the residual plots that I drew on the board:
These ones here are fine, note the lack of relationship between e and x
That one is no good because there is a relationship between x and e, basically, error increases as x increases, not really that random is it?
This one shows a group of outliers in the top right. So, again, we have a problem.
Listen Now
[tags]Dave Brodbeck, Psychology, Statistics, Algoma University, Regression[/tags]