Root Mean Square Error(RMSE)

\[RMSE=\sqrt{\frac{\sum_{i=0}^n (\widehat{y_i}-y_i)^2}{n}}\]

The RMSE is an index that provides an indication of the dispersion or variability in the quality of the prediction. The RMSE can be related to the variance of the model. Often, RMSE values are difficult to interpret because it is not possible to tell whether a variance value is low or high. To overcome this, it is more interesting to normalize the RMSE so that this indicator is expressed as a percentage of the mean value of the observations. This can be used to make the indicator more meaningful.