What exactly does it mean to "regress out" a variable or quantity? This is something I hear about all the time, but it's strangely difficult to find a precise description online. Any pointers - especially to algorithmic descriptions - would be much appreciated.
-
1$\begingroup$ Can you point to an example (or even better, multiple examples) of this phrase being used within context? $\endgroup$– Jake WestfallSep 21, 2014 at 22:10
-
$\begingroup$ @JakeWestfall Yes: here are two earlier questions on Cross Validated: stats.stackexchange.com/questions/5713/… stats.stackexchange.com/questions/5713/… $\endgroup$– HansSchwabingSep 21, 2014 at 22:16
1 Answer
Although it is not the most commonly used phrase (in my experience), as used in the question you linked to, to "regress out" a third variable is synonymous with "partialling out," "controlling for," "adjusting for," or "while holding constant the value of" another predictor variable. These concepts have been explained at various places on this site; see for example the collection of links in this answer.
-
3$\begingroup$ I would add "adjusting for" and "while holding the value of variable constant" to your list of synonyms. $\endgroup$– AlexisSep 21, 2014 at 23:07