Dear Statalisters,
I need to test for the endogeneity of a discrete RHS
variable(smoker) in a probit regression. so i have my
"main" equation as:
probit highbp smoker smokyrs age age2 sex alcohol
socclass asian black salt underw normal overw sports
where highbp takes the value 0 or 1 and smoker also
takes 0 or 1.
I want to test if smoker is endogenous and my
instruments are january-december. so i regressed my
endogenous variable on the instruments+other exogenous
variables and generated the generalised residuals
called smokerhat
probit smoker smokyrs jan feb march apr may june july
aug sept oct nov age age2 alcohol sex socclass asian
black salt underw normal overw sports,
score(smokerhat)
then I plugged the generalised residuals(Smokerhat)
into my 'main' equation:
probit highbp smoker smokyrs smokerhat age age2 sex
alcohol socclass asian black salt underw normal overw
sports
i got the following:
Coeff std error z
p>|z|
smokerhat | .0296293 .0461569 0.64 0.521 ,
My question however is whether the durbin-wu-hausman
test can actually be applied to IVprobit and whether
the procedure i folllowed is correct or not?
Secondly, if i run a simple ivprobit as follows:
ivprobit highbp smokyrs (smoker=jan feb march apr may
june july aug sept oct nov) age age2 alcohol sex
socclass asian black salt underw normal overw sports
I get the wald test of exogeneity :
Wald test of exogeneity (/athrho = 0): chi2(1) =
0.51 Prob > chi2 = 0.4742
what does that mean?my smoker variable is endogenous
or exogenous?
thank you for your help.
Vicky
__________________________________________________________
Sent from Yahoo! Mail.
A Smarter Inbox. http://uk.docs.yahoo.com/nowyoucan.html
*
* For searches and help try:
* http://www.stata.com/support/faqs/res/findit.html
* http://www.stata.com/support/statalist/faq
* http://www.ats.ucla.edu/stat/stata/
