Generally speaking, t-test is used for testing the pearson correlation coefficient significance of X's and Y's. In my case, however, assuming X and Y both follow some known distributions. I have a series of data (x1, y1), (x2, y2),..., (xn, yn) drawn from X and Y. If I want to test the significance of the pearson correlation coefficient of X's and Y's, should I use z-test since distribution is known or use t-test since it is the default?


1 Answer 1


Testing the significance of a correlation is akin to performing a regression of y on x and evaluating the coefficient of x in the model. The reason you use a t-test in that scenario is because we don't know the variance of $y\vert x$; It must be estimated.

I would suppose that if you knew the variance of $y \vert x$ a priori , you could use a z test.

But honestly, if $n$ is large enough, there will not be a huge difference between the two.


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge that you have read and understand our privacy policy and code of conduct.

Not the answer you're looking for? Browse other questions tagged or ask your own question.