If I have a regression or classification problem

y ~ x1 + x2 + ... + xn

I might call the xi the regressors or the predictors. But what do you call y?

The terms regressand and predictand are clunky, don't seem particularly standard, and have the disadvantage that they're very similar to the words for the right-hand-side variables (I'm going to be using these terms in type declarations and documentation for a program, so minimizing the potential for confusion is a priority).

Also, using regressor and regressand is inappropriate when you have a classification problem, rather than a regression.

Is there a commonly used term?

  • 6
    $\begingroup$ I think the "response" or "outcome" variable are commonly used terms. $\endgroup$
    – Macro
    Aug 28, 2012 at 15:27
  • $\begingroup$ What if you are using regression to do classification? $\endgroup$
    – jkd
    Aug 28, 2012 at 15:56
  • $\begingroup$ 'outcome' also seems fine for classification. Or try 'class' or 'category'. If you're really fitting regression models for classification purposes you should probably stop, and use a proper model instead ;-) $\endgroup$ Aug 28, 2012 at 16:11
  • 3
    $\begingroup$ In regression it is also called the dependent variable or DV and the Xs are called the independent variables or IVs. However the independent variables do not have to be and usually are not statistical independent. $\endgroup$ Aug 28, 2012 at 16:14
  • $\begingroup$ @ConjugatePrior Is logistic regression considered a regression method? Since glm is a generalization of the linear model, cannot glm be considered regression too? $\endgroup$
    – jkd
    Aug 28, 2012 at 16:44

1 Answer 1


The "response" or "outcome" variables are commonly used terms.

  • $\begingroup$ Some fields talk of "criterion" variables. $\endgroup$
    – Nick Cox
    Feb 11, 2014 at 21:28

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.