1
$\begingroup$

in the textbook "Introductory to Econometrics" Wooldridge writes

$\hat{\beta}_1=\hat{p}_{xy}\cdot(\frac{\hat{\sigma}_x}{\hat{\sigma}_y})$

where $\hat{\beta}_1$ is the OLS-Estimator, $\hat{p}_{xy}$ is the correlation between $x$ and $y$, and $\hat{\sigma}_x$ is the standard deviation of $x$. When I derive this formula I always come up with

$\hat{\beta}_1=\hat{p}_{xy}\cdot(\frac{\hat{\sigma}_y}{\hat{\sigma}_x})$.

Also according to wikipedia this formula is true. Did Wooldridge accidentally invert the fraction?

$\endgroup$
1
  • 2
    $\begingroup$ It looks like you got it right. Wooldridge made a mistake at least with that one case. $\endgroup$ Mar 29, 2017 at 22:56

1 Answer 1

2
$\begingroup$

Yes this is a typo. In simple regression,

$$y = a + bx + u$$

the OLS estimator for $b$ is $$\hat b = \frac {\hat Cov(x,y)}{\hat \sigma^2_x}$$

Since $\text{Cov}(x,y) = \hat \rho_{x,y}\cdot \hat \sigma_x \cdot \hat \sigma_y$ we get

$$\hat b = \hat \rho_{x,y}\frac {\hat \sigma_y}{\hat \sigma_x}$$

$\endgroup$

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.