Back to the Top
Hello,
I am trying to test the bias and precision of an equation.
I would like to know the relevance of Mean Square Error and the Root
Mean Square Error.
Questions I have are
1) Is there a limit (like 15% or 20% for example) that can be
applied to these values?
2) Which error (Mean square error or the root mean square error)
should be reported in a publication?
Relevant and important insight is requested
Thank you
Anisha Mendonza
PhD candidate
University of Rhode Island
Back to the Top
The following message was posted to: PharmPK
Dear Anisha,
In reply to your questions:
1) There is no 'general limit' applicable to MSE and RMSE. These
measures of
precision (or better: imprecision) should of course be as low as
possible.
For some applications, e.g. predictability of drug response, 15% or
20% may
be acceptable, but this is arbitrary.
2) RMSE expressed as a percentage of the 'true value' or 'observed
value' is
generally most imformative and therefore preferred for publication. MSE
cannot be expressed as a percentage, and its value is not easily
interpreted.
Please note that the measure of bias is Mean Error (ME). This value,
also
expressed as a percentage, may be reported together with RMSE.
Best regards,
Hans Proost
Johannes H. Proost
Dept. of Pharmacokinetics and Drug Delivery
University Centre for Pharmacy
Antonius Deusinglaan 1
9713 AV Groningen, The Netherlands
tel. 31-50 363 3292
fax 31-50 363 3247
Email: j.h.proost.at.rug.nl
PharmPK Discussion List Archive Index page
Copyright 1995-2010 David W. A. Bourne (david@boomer.org)