0
$\begingroup$

We are considering a simple linear model $Y = \beta_0 + \beta_1X + \epsilon$. I am given the following quantities. $\bar{X}$, $\bar{Y}$, $s_x$, $s_y$, $\rho_{xy}$, and $n$ the sample size. I am asked to compute all of the following: $\hat{\beta_0}$, $\hat{\beta_1}$, $s_{\beta_1}$. $s_{\beta_0}$, and all the values in the ANOVA table(SSR, SSE, SSTO). I understand how to compute the values for the least squares estimators, but I dont understand how to compute their standard deviations or compute the ANOVA statistics without the actual values.

$\endgroup$
0
$\begingroup$

See Fitting the regression line section on Wiki on simple Linear regression for the first two. See the Normality assumption section on the same page for the latter two.

For the Anova table figures, use that $R^2$ is $\rho_{xy}^2$, recall the connection between $(s_y, n)$ and SSTO and the connection between SSR, SSE, SSTO and the $R^2$.

$\endgroup$

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.