Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. Join them; it only takes a minute:

Sign up
Here's how it works:
  1. Anybody can ask a question
  2. Anybody can answer
  3. The best answers are voted up and rise to the top

I came through this equation and would like to learn more about how to inteprete it. d is defined as actual value, and $\hat{d}$ is defined as predicted value.

Why does this equation (a) divides sum of least square error to the sum of $d_{i,j}^2$ ? (b) takes square root of (a) ?

May I know what's the name of this equation, that I can read further...

$$ \sqrt{\frac{\sum_{i,j} (d_{i,j} - \hat{d}_{i,j})^2}{\sum_{i,j} d_{i,j}^2}} $$

share|cite|improve this question
"I came through this equation..." - where? Surely there would be some sort of explanation in the book/paper where you saw it, or a pointer to where this is explained. – J. M. Aug 14 '12 at 10:00
Hi, it's from Section 4 of this paper… – twfx Aug 14 '12 at 10:56
up vote 0 down vote accepted

a) it is a normalization factor. Assume that you have perfect prediction. Then you will get $0$. Say your predictor is restricted to a closed interval $[0,1]$ and your estimates are all zeros and actual values are all ones. Then you will get $1$. b)square root is to make it as a distance in $N$ dimensional Euclidean space, such as $z=\sqrt{(x_1-y_1)^2+(x_2-y_2)^2+...}$

share|cite|improve this answer

Your Answer


By posting your answer, you agree to the privacy policy and terms of service.

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