Clarify operates under the assumption that the model you specified is
correct. If it is, then you can hold constant or change any variable
or set of variables as you see fit, and then observe the consequence
of these settings for your predicted or expected values. so that
means that if you want, you can compute any quantity you desire. you
can hold intermediate variables constant and look at how a change in a
pre-treatment variable affects the outcome, or anything else.
Gary
---
http://gking.harvard.edu
On Fri, Apr 23, 2010 at 3:27 PM, Thomas Didier
<thomas.didier(a)mail.mcgill.ca> wrote:
Dear Clarify listers,
I have a model including indirect effects:
A. y1=f(x1, x2, y2, y3)
B. y2=f(x1, x2)
C. y3=f(x1, x2)
I am using the CLARIFY plugin in Stata to predict the effect of x1 on y1. Clarify is very
useful to predict the outcome of each equation individually with the simqi command, but as
far as I understand it this command does not assess the role of indirect effects.
Can Clarify make a prediction considering the individual direct effect of x1 on y1 as
well as the indirect effect of x1 on y1 through y2 and y3? Can it also calculate the total
effect of x1 on y1?
Any help in this regard would be greatly appreciated.
Best regards,
Thomas
McGill University
Thomas.Didier@mail.mcgill.ca<mailto:Thomas.Didier@mail.mcgill.ca>
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=clarify
--
Clarify mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=clarify