0
$\begingroup$

I am tasked with giving an example of a dataset in which the presence of an outlier dramatically affects the performance of Ordinary Least Squares (OLS) regression. I've searched and searched the web and I understand that OLS has a hard time dealing with outliers, but I'm having a hard time figuring out why, and finding a dataset to prove this.

$\endgroup$
0
$\begingroup$

If you are looking for a real-world data set here is one on Harvard's Dataverse that examines state social politics research for outliers for the same purpose you are looking for. If you are looking for one for more illustrative purposes one data set worth knowing is Anscombe's quartet for demonstrating how misleading some descriptive statistics can be. For your own investigations, many data sets with and without outliers can be found on Google's beta dataset search and are worth exploring if you are curious!

$\endgroup$
1
  • $\begingroup$ I will definitely look into those. Especially the quartet is interesting to me. Thanks for the help, It is much appreciated! $\endgroup$ Nov 6 '18 at 14:54

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.