However, when I try to do this, I find that my manually run first-stage
regression doesn't match was Stata reports when doing the ivreg. I've
made sure I'm running the command on the same sample. (See below).
So, hopefully someone can come up with a pre-canned way to include both
the first and second stage regressions in outreg, etc. Failing that, can
anyone explain why I'm getting different results when I run the
regression on lawyers_pc manually versus when it is run in the first
stage of ivreg?
Thanks.
Glenn
.ivreg common_action_new_pc refaffect (lawyers_pc =passed_bar ), first
First-stage regressions
-----------------------
-----------------------------------------------
lawyers_pc | Coef. Std. Err. t
-------------+---------------------------------
refaffect | 1.77e-06 .0000127 0.14
passed_bar | 2.55e-08 3.18e-08 0.80
_cons | .0000509 .00002 2.55
-----------------------------------------------
Instrumental variables (2SLS) regression
[omitted]
-----------------------------------------------
Instrumented: lawyers_pc
Instruments: refaffect passed_bar >>>>CONFIRMS WHAT IS INSTRUMENTED BY
WHAT
-----------------------------------------------
.reg lawyers_pc refaffect passed_bar >>>>>>>THE MANUAL COMMAND
-----------------------------------------------
lawyers_pc | Coef. Std. Err. t
-------------+---------------------------------
refaffect | 1.10e-06 .0000123 0.09
passed_bar | 2.87e-08 2.91e-08 0.98
_cons | .0000486 .0000179 2.72
-----------------------------------------------
Glenn Hoetker
Assistant Professor of Strategy
College of Business
University of Illinois at Urbana-Champaign
217-265-4081
[email protected]
*
* 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/
