Statistical Glossary
RMSE:
RMSE is root mean squared error. In predicting a numerical outcome with a statistical model, predicted values rarely match actual outcomes exactly. The difference between predicted and actual is the error (or residual). To calculate RMSE, square each error, take the average, then take the square root. The result is what you might think of as the "typical error" for a model. Squaring ensures that errors in one direction do not offset errors in the other direction..