can anyone give me the proof that the RMS error for the regression of y on x will always be the square root of 1 minus the correlation of y on x, times the SD of y? I know the intuition behind it. like when the r value is equal to 1 the rmse is 0 cuz all points lie on a straight line, and when r is zero then the rmse is the same as the sd because there is no correlation between the x and v values. what I want is the algebraic proof of the relationship


1 Answer 1


I know its too late, but if someone stumbles on this question here is the answer.




  • $\begingroup$ If these links go dead your answer will be useless. Can you edit it to include the proofs which the OP asked for? $\endgroup$
    – mdewey
    Commented Jan 20, 2019 at 15:22

Your Answer

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

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