Dear all,
I am trying to estimate a Y=bX equation where the x are 4 endogenous ordered
categories. in order to control for endogeneity I run an ordered probit on
X=dZ, estimate the inverse mill's ratios and plug them in the main equation.
from the model I have 4 Mill's ratios (L).
y= b1 X1 + b2 X2 + b3 X3 + c0 L0 + c1 L1 + c2 L2 + c3 L3
I managed to estimate the equation using stata (I wrote a small programme) but
my problem is to correct the standard errors to test the parameters.
stata 7.0 does the job for the probit selection and not when an ordered probit
is used. limdep does the job when you just have one mill's ratio and an
ordered probit, i.e. you have to estimate one main equation for each category.
(this is also what the paper I found on the internet did)
in my case instead everything is in one equation.
do you know of any paper,program.. that solve this problem?
thanks
Mario Fiorini
PhD student UCL
*
* 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/