The difference between confidence and prediction intervals has been explained in plain English on Cross Validated: Difference between confidence intervals and prediction intervals

However, when should I use which measure? Are there situations where both measures should be used?

I am looking for concrete examples illustrating when to pick either of them or both.

  • $\begingroup$ When you should use which measure? It depends on what you want to do. As the names suggest, if your goal is prediction of y, you use the prediction interval. IF your goal is to gauge how confident you should be in your estimator, you "constrain" your prediction by x. That's why you're calculating the expected value of y conditional/given x. $\endgroup$ – rocinante Apr 5 '14 at 12:36
  • $\begingroup$ If you want to guess the value of an individual with given properties, then look at the prediction interval. If you want to guess the average of all individuals in the population having these properties, check the confidence interval. $\endgroup$ – Michael M May 17 '14 at 21:23

Your Answer

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

Browse other questions tagged or ask your own question.