0
$\begingroup$

Here's an example:

If I estimate a model WAGE = b1 + b2*EDUC + b3*EXPER and we assume that EDUC (years of education) is endogenous while EXPER (years of work exerpience) is exogenous, will my estimate for b3 and b1 be just fine or does the bias from EDUC spill over into the other terms?

Thank you.

$\endgroup$
1
$\begingroup$

In general terms if there are good reason for think that some estimated parameters are biased, even only one, then also the others are biased. Orthogonality conditions among regressors can produce exceptions.

$\endgroup$

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.