2
$\begingroup$

Generally, both of them can represent the linear relationship between x and y scale to [0,1]. Are they 99% very similar?

$\endgroup$
  • $\begingroup$ Not the same thing. See answers. $\endgroup$ – BruceET Jun 19 at 5:41
4
$\begingroup$

Computationally, R-sq in computer printouts is the square $r^2$of the (Pearson) correlation $r.$ Sometimes $r^2$ is called the 'coefficient of determination'. Because $-1 \le r \le 1$ we have $0 \le r^2 \le 1.$

Suppose we have the simple linear regression model $$Y_i = \beta_0 + \beta_1 x_i + e_i,$$ for $i = 1,2, \dots,n$ where $e_i \stackrel{iid}{\sim}\mathsf{Norm}(0, \sigma_e).$

In regression $r^2$ is sometimes referred to (intuitively) as the proportion of the variability in $Y$ that is explained by regression on $x.$ This interpretation roughly matches the equation $$S_{Y|x}^2 = \frac{n-1}{n-2}S_Y^2(1 - r^2),$$ where $S_Y^2$ is the sample variance of the $Y_i$ and $S_{Y|x}^2$ is the sum of squared residuals divided by $n-2,$ sometimes referred to as the variance about the regression line.

Thus if $r = \pm 1$ so that $r^2 = 1,$ then $S_{Y|x}^2 = 0$ and all of the $(x_i,Y_i)$-points lie precisely on the regression line. Also, if $r \approx 0,$ then $S_{Y|x}^2 \approx S_Y^2$ and the $x_i$ have no role to play in 'explaining' the $Y_i.$

Sometimes, the Pearson 'correlation coefficient' $r$ is said to express the 'linear component' of the association of $X_i$ and $Y_i.$

$\endgroup$
  • 1
    $\begingroup$ @mkt. Thanks for fixing typo. $\endgroup$ – BruceET Jun 19 at 5:59
1
$\begingroup$

R-squared or coefficient of determination generally has the value between 0 and 1. However cor(x,y) is between -1 and 1. Definition of R-squared for a multiple linear regression is the square of correlation between output (y) and predicted values(f). For the case of simple linear regression with an intercept and one explanatory variable it happens R-squared is as same as square of cor(x,y). I hope this helps you.

$\endgroup$
  • $\begingroup$ Definitely they are not same, but for simple linear regression : Y=ax+b (x is a single variable and not a vector), value of coefficient of determination is equal to square of correlation between x and y. $\endgroup$ – Masoud Norouzi Darabad Jun 19 at 5:19

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

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