I recently discovered my colleague (not a mathematician) was evaluating their experimental regression analyses by reporting the mean percent difference of each estimated output (from their fitted curve) with the true output.

My instincts would be to evaluate goodness of fit with root mean squared error.

Is my colleague's method of evaluating fitted curves superior, or are there reasons supporting RMSE?


Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Browse other questions tagged or ask your own question.