Statalist The Stata Listserver


[Date Prev][Date Next][Thread Prev][Thread Next][Date index][Thread index]

st: Re: 2SLS with the first stage being a probit.


From   "Rodrigo A. Alfaro" <[email protected]>
To   <[email protected]>
Subject   st: Re: 2SLS with the first stage being a probit.
Date   Mon, 9 Oct 2006 23:38:35 -0400

Jackie,

Mark suggested you -treatreg- for your previous post. It seems that
this is not what you want...

Pretend x is your endogenous variable and x_f is the forecast after probit.

(1) Note that x_f was obtained by other procedure then x_f may have less
observations than x. Take a look with -sum x x_f-
(2) Wooldridge (2002) suggests (procedure 18.1) -reg y x (x_f)- instead
of -reg y x_f-... the latter sounds what you are doing.
(3) If you still think that -reg y x_f- is what you want then you should
run -reg y x_f if x!=.-.
(4) You always can create x_f from another model such as logit or whatever
do it to check that results are robust, of course under procedure 18.1
assumptions.

Rodrigo.


----- Original Message ----- 
From: "Hui Wang" <[email protected]>
To: <[email protected]>
Sent: Monday, October 09, 2006 10:52 PM
Subject: st: 2SLS with the first stage being a probit.


Hi all,

I'd like to know that whether there is any command in stata allowing me to
conduct a 2SLS regression with the 1st stage being a probit regression
regressing endogenous variable on instruments.

I tried to program the above 2SLS. First do the probit and get the predict
value and then use the predicted value to replace the endogenous variable in
the 2nd stage regression. But the problem is that the observation number of
the two stages are not the same. This is not the case when I use the command
like 'ivreg'.  If I have to write a program to conduct the 2SLS mentioned
above, how should I do in order to get equal observation in the two stage
regressions?

Thank you.

Jackie

_________________________________________________________________
Express yourself instantly with MSN Messenger! Download today it's FREE!
http://messenger.msn.click-url.com/go/onm00200471ave/direct/01/

*
*   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/

*
*   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/



© Copyright 1996–2024 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   What's new   |   Site index