I am doing some oaxaca decompositions and I'm looking for some shortcuts,
thus I ran the two following estimations:
[1] separate estimations for male and female: wage = alpha + beta1*gamma +
beta2*zeta
[2] one estimation, but with interactive dummies (female = 1 for women)
wage = alpha + beta1*gamma + beta2*zeta + beta3*female +
beta4*female*gamma + beta5*female*zeta
Now my question is why is it that the estimated beta1 in [2] differ from
that of beta1 in the male estimation in [1]? And why is it that the
summation of beta1 and beta4 in estimation [2] would equal to beta1 for
females in estimation [1]?
I know I'm making different assumptions regarding the error terms in
estimations [1] and [2], but what elses are driving the differences?
Many thanks in advance.
Cheers,
Ada Ma
-----------------------------------------
Dept of Econ
Univ of Aberdeen [email protected]
_________________________________________________________________
Stay in touch with absent friends - get MSN Messenger
http://www.msn.co.uk/messenger
