3
$\begingroup$

By going through the method in which we reject a null hypothesis: any observed sample mean value which falls in the rejection region; that is when test statistic > Z(1-alpha) (consider > alternate hypothesis). The definition of Power is the probability of getting an extreme (=alpha) or even more than that. I infer that both refers to the same region and hence their area is nothing different and thus Power is the probability of rejecting a null hypothesis when it is FALSE. Please let me know if I am missing anything. Help much appreciated.

PS: I dont have a statistical background

$\endgroup$

migrated from stackoverflow.com Feb 19 '17 at 16:49

This question came from our site for professional and enthusiast programmers.

10
$\begingroup$

Power is the probability of rejecting a null hypothesis when it is FALSE

This is correct; specifically at some particular point in the alternative (each has its own power).

By contrast, the rejection region different; indeed it is not even a probability, it's a subset of the values that the test statistic can take (in particular, the subset for which you will reject the null hypothesis).

Power is the probability of getting a test statistic in the rejection region, at some particular alternative (some point in the space of the alternative). If the alternatives are indexed by a parameter you can find the power as a function of it (obtaining a power curve). The rejection region is the same at each point but the power would be different.

$\endgroup$

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.