Dear Professor Parker

You can take a look at the following link first.

http://www2.chass.ncsu.edu/garson/pa765/2sls.htm

It might solve most questions you asked.
If not, my suggestion is writting down the latent variables and covariates 
you have and the coefficients you got from stata, and then you can figure 
the substantive effects out.

Regards

Alex Chang
Ph.D, University of Iowa 2007.


>From: "Parker, David C W" <[log in to unmask]>
>Reply-To: Political Methodology Society <[log in to unmask]>
>To: [log in to unmask]
>Subject: [POLMETH] IVREG QUestion
>Date: Thu, 10 May 2007 22:06:05 -0400
>
>Hello--
>
>I've run a system of equations using Stata's IVREG procedure. I'm 
interested in interpreting the results of instrumented variable in the 
model. As the variable is instrumented, I assume for the purposes of 
calculating the substantive effect of the coefficient I need to know the 
distribution of the instrumented variable as opposed to the original 
variable, yes? For example, the original variable ranges from -5 to 4. But 
I assume the instrumented variable has some other range. And if so, how do 
I figure out that distribution so I can interpret the coefficient 
substantively?
>
>Any help would be greatly appreciated.
>
>David C.W. Parker
>Assistant Professor
>Department of Political Science
>Indiana University South Bend
>P.O. Box 7111
>South Bend, IN 46634
>Phone: 574.520.4817

_________________________________________________________________
Windows Live Messenger 正式版熱烈下載中：共用資料夾，傳檔案再大也不怕 
http://get.live.com/messenger/overview