Regression - Expected Values and Variances of the OLS estimators

4 important questions on Regression - Expected Values and Variances of the OLS estimators

What four assumptions (SLR1-SLR4) are needed for unbiasedness of OLS?

SLR1 (Linear in parameters): y = B0 + B1x
SLR2 (Random sampling): We have random sample {(x(i),y(i));(i)=1,..,n} following SLR1's population model
SLR3 (Sample variation in the explanatory variable): The sample values x1,..,xn of the explanatory variable are not all the same
SLR4 (Zero conditional mean): E(u|x)=0
SLR1-SLR4 gives us E(^B0)=B0 and E(^B1)=B1

What additional assumption is needed for deriving variances of OLS estimators?

Denote the error variance var(u) with o2

SLR5 (Homoskedasticity): var(u|x) = o2
SLR4 and SLR5 are equivalent to E[y|x]=B0+B1x and var(y|x)=o2

What is the sampling variance under SLR1-SLR5?

It's the same, but now it's divided by the total sum of squares. 
  • Higher grades + faster learning
  • Never study anything twice
  • 100% sure, 100% understanding
Discover Study Smart

How can the error variance and the standard error be estimated?

The error variance o2 can be estimated with ô2 = SSR / n-2. Under assumptions SLR1-SLR5, E(ô2) = o2

The question on the page originate from the summary of the following study material:

  • A unique study and practice tool
  • Never study anything twice again
  • Get the grades you hope for
  • 100% sure, 100% understanding
Remember faster, study better. Scientifically proven.
Trustpilot Logo