Question-and-Answer Resource for the Building Energy Modeling Community
Get started with the Help page
Ask Your Question

CalibrationReport CV-RMSE and NMBE computation

asked 2021-11-16 15:20:58 -0600

rimas-maalka's avatar

updated 2022-02-07 14:20:57 -0600

In the UtilityBill measure, the formulas for computing NMBE and CV-RMSE include an 'n-1' rather than just an 'n' in the denominator. For example:


result = 100.0 * std::pow(squaredError / (n - 1), 0.5) / ybar;
result = 100.0 * (sumError / (n - 1)) / ybar;

Why doesn't the formula just use 'n' as described in the definition, e.g.

Thank you

edit retag flag offensive close merge delete

1 Answer

Sort by ยป oldest newest most voted

answered 2021-11-16 20:15:45 -0600

In this case, when you are estimating the variance you don't know what the real mean is so you have to use an estimate of the mean instead of the true mean. Estimating the mean reduces the degrees of freedom (number of samples) by one so you use N-1 instead of N which results in a slightly larger estimate of the variance. This link might help a bit:

edit flag offensive delete link more

Your Answer

Please start posting anonymously - your entry will be published after you log in or create a new account.

Add Answer


Question Tools


Asked: 2021-11-16 15:20:58 -0600

Seen: 176 times

Last updated: Nov 17 '21