2
$\begingroup$

Whereby one method is to square the $r$ (hence $r^2$) or the correlation coefficient, and the other method is 1 - (Sum of Squared Errors (from the regression line) / Sum of Squared Errors (from the mean y), i.e. $$R^2=1-\frac{\text{SSE}}{\text{SST}}$$

$\endgroup$
1

1 Answer 1

2
$\begingroup$

In simple linear regression, we have $\hat{y_i}=\hat{\beta}x_i+\hat{\alpha}$, where $$\hat{\beta}={{\sum (x_i-\bar{x})(y_i-\bar{y})}\over {\sum(x_i-\bar{x})^2}},\ \ \ \hat{\alpha}=\bar{y}-\hat{\beta}\bar{x}\rightarrow \hat{y}_i-\bar{y}=\hat{\beta}(x_i-\bar{x})$$ Substituting this into the formula of coefficient of determination: $$\begin{align}R^2&=\frac{SSR}{SST}=\frac{\sum (\hat{y_i}-\bar{y})^2}{\sum (y_i-\bar{y})^2}=\hat{\beta}^2\frac{\sum (x_i-\bar{x})^2}{\sum (y_i-\bar{y})^2}\\&=\left({{\sum (x_i-\bar{x})(y_i-\bar{y})}\over {\sum(x_i-\bar{x})^2}}\right)^2\frac{\sum (x_i-\bar{x})^2}{\sum (y_i-\bar{y})^2}\\&=\left(\underbrace{{{\sum (x_i-\bar{x})(y_i-\bar{y})}\over {\sqrt{\sum(x_i-\bar{x})^2\sum (y_i-\bar{y})^2}}}}_{R=\text{correlation coef.}}\right)^2\end{align}$$

$\endgroup$

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.