Residual standard deviation is a measure of how far the observed values in a regression lie from the values the fitted model predicts. Also called the standard error of the estimate or standard error of the regression, it is the standard deviation of the residuals, adjusted for the number of parameters estimated:
Here is an observed value, the fitted value, the number of observations and the number of estimated parameters including the intercept, so in a simple linear regression. For example, a simple regression on 11 observations whose squared residuals sum to 18 gives .
The figure is in the same units as the dependent variable, which makes it easier to read than R-squared; a smaller value means a tighter fit. In finance it is used to judge factor and multivariate regression models, and in a market model regression of a share's returns it measures idiosyncratic, or specific, risk.
Ref 1100001931