# Which performance measure to use when using SVM: MSE or MAE?

It is a common practice to measure an SVM model's performance by calculating its MSE (Mean Square Error). Why not use Mean Absolute Error (averaging errors' absolute values instead of squared values)?

-
Thank you all for the useful answers. –  Bliss Jul 26 '12 at 12:58
Yan, I've merged your two accounts. Please, register this one to benefit from Stack Exchange facilities, such as voting, system notifications, commenting, etc. (BTW, it would be good to review your past questions and check whether any of their associated replies deserve some votes or green check--i.e., if one answer solved your problem, mark it as such so that future users will know that it was helpful.) –  chl Jul 26 '12 at 13:13

Actually, looking at both MAE and RMSE gives you additional information about the distribution of the errors:

$\mathrm{MAE} \leq \mathrm{RMSE} \leq \mathrm{MAE}^2$ (for regression)

• if $\mathrm{RMSE}$ is close to $\mathrm{MAE}$, the model makes many relatively small errors
• if $\mathrm{RMSE}$ is close to $\mathrm{MAE}^2$, the model makes few but large errors
-

MAE is more intuitive than MSE to simply evaluate the overall error.

MSE is easier to handle mathematically for variance analysis. For example, MSE is used to calculate the error variance $s_e^2$, which is a recurring value in regression statistics.

-