Bookmark and Share

Notice: On April 23, 2014, Statalist moved from an email list to a forum, based at statalist.org.


[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

st: Running SUR for multiple IV regressions


From   Tomoo Okubo <tomoo.okubo@gmail.com>
To   statalist <statalist@hsphsun2.harvard.edu>
Subject   st: Running SUR for multiple IV regressions
Date   Wed, 31 Jul 2013 09:30:48 -0400

Hi everyone,

I am trying to analyze mean standardized effect size of across
multiple outcomes (outcome1 outcome2), following Kling et al. (2004)
and Clingingsmith et al. (Q J Econ, 2009).
Stata command "avg_effect" allows to standardize multiple outcomes
using the spread of the outcome for the control group and compute the
average effect size for a group of outcomes in a SUR framework to take
into account the covariance between the different outcome variables.
However, it DOES NOT ALLOW TO RUN SUR ON MULTIPLE IV REGRESSION to
compute the effect size, and was wondering if it is possible to write
a code to make this possible.
I am copying the code I have tried so far, but STATA DOES NOT RUN SUR,
SAYING THAT IT DOES NOT SUPPORT IVREGRESS.

I appreciate any ideas or help on this. Thank you very much for your time!

est clear
ivregress 2sls outcome1 control (compliance_measure = treatment control)
estimates store one
ivregress 2sls outcome2 control (compliance_measure = treatment control)
estimates store two
preserve

qui sum outcome1 if treatment==0
local sd_1=r(sd)
qui sum outcome2 if treatment==0
local sd_2=r(sd)

suest one two, svy
*sur does not support -ivreg-
ereturn list
mat b=e(b)
mat list b

local
ae_treatment="(1/2)*(1/`sd_1')*[a_mean]treatment+(1/2)*(1/`sd_2')*[b_mean]treatment"
nlcom `ae_treatment', post

Best,

Tomoo Okubo
Harvard Kennedy School
*
*   For searches and help try:
*   http://www.stata.com/help.cgi?search
*   http://www.stata.com/support/faqs/resources/statalist-faq/
*   http://www.ats.ucla.edu/stat/stata/


© Copyright 1996–2018 StataCorp LLC   |   Terms of use   |   Privacy   |   Contact us   |   Site index