do i need to report all the various interesting ways in which my data violate parametric assumptions? like surely one violated assumption should be enough, right?
Doesn't it depend on what do you want to do with the model? People use linear regressions all the time, even though few real-world variables are normally distributed.
![[personal profile]](https://www.dreamwidth.org/img/silk/identity/user.png)
timeasmymeasure
no subject
Date: 2020-06-23 06:56 pm (UTC)