Would a simple squared covariate actually work? Negative values should be treated differently, right? Then wouldn't cubing make more sense?

When checking for non-linear relationships, squaring a covariate seems to be common. But wouldn't cubing make more sense to preserve negative values?

This topic was automatically closed 21 days after the last reply. New replies are no longer allowed.

If you have a query related to it or one of the replies, start a new topic and refer back with a link.