[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]
st: Re: IVREG or IVREG 2
I just subscribed, to statlist, I do not know if the earlier one went
through of not. Please accept my apologies if you are seeing this for
the second time.
I will describe the problem i have, then my question
I have a classical labor supply model addressing wage is endogenous.
I have looked into http://www.stata.com/support/faqs/stat/ivreg.html
But this is not addressing the problem, I have.
ln wage= aX + bZ + e
where z is the instrumental variable, then I have the second stage
which is the labor supply equation
ln weekly hrs = eX + klnwagehat + E
where lnwagehat is the predicted wage adjusted with standard errors,
so wage in stata equals to the following:
wagepathat = exp(predicted mean of wage from 1st stage + SE/2), where
SE is standard error
If I not mistaken, by using ivreg or ivreg2 the predicted wage is
not corrected for. Is there anyway I can correct for the standard
error in ivreg and ivreg2 so that the predicted value will account for
the standard error and I can use for example ivreg2 or ivendog for
Wu-Hausman test, and other tests.
* For searches and help try: