2
$\begingroup$

My question directly relates to one that has already been answered - here: Expected Value and Variance of Estimation of Slope Parameter $\beta_1$ in Simple Linear Regression

However, there is an initial step in the derivation which I don't understand which the proof and the textbook I am reading also just assumes.

The equation for $\hat{\beta_1}$ as a result from the normal equations is:
$\hat{\beta_1}=\sum_i(x_i−\bar{x})(Y_i−\bar{Y})/\sum_i(x_i−\bar{x})^2$

However, when we get to deriving the variance of $\hat{\beta_1}$ we get to this expression (written better on the previous question above):

$\hat{β}_1=\sum_i(x_i−\bar{x})(Y_i−\bar{Y})/\sum_i(x_i−\bar{x})^2=\sum_i(x_i−\bar{x})Y_i/S_{xx}$

My main question is what happened to the $\bar{Y}$ in the numerator? What is happening here?

$\endgroup$
3
$\begingroup$

Consider: $\sum_i(x_i−\bar{x})(Y_i−\bar{Y})$

  • Expand the second term in the product so you can organize it into a difference of two sums.

  • take multiplicative constants (/things that don't change with $i$) outside the sum in the 2nd term

  • show that the second term is zero

| cite | improve this answer | |
$\endgroup$

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.