# R^2 coefficient of determination and coefficient of correlation

For every model of relationship between y and x

If all the observations are on the model then r^2=1

Fortunately r turns out to be normalized covariance    🙂

http://maths.nayland.school.nz/Year_13_Maths/3.9_Bivariate_data/7_Coefficient_of_determination.html

$\bar R^2 = {1-{SS_\text{res}/df_e \over SS_\text{tot}/df_t}}$
$R^{2} = {1-{\textit{VAR}_\text{err} \over \textit{VAR}_\text{tot}}}$