Yeah, that's what I was looking for. Thank you very much.
On Jul 9, 2008, at 1:18 PM, Gary King wrote:
as I recall, there's some R code in the
replication data set for
the matchit article that does this. i'd have a look there. you
can find it on my web page under dataverse.
and by the way, if anyone makes data available, you're all welcome
to have your own dataverse too. see
http://thedata.org.
Gary
---
http://gking.harvard.edu
Keith Schnakenberg wrote:
Hello former classmates,
At the end of the notes from the matching lecture, There are two
graphics (one related to Carpenter's article on FDA drug approval
and the other related to Koch's 2002 article on woman legislators)
in which someone estimates the ATT for some treatment across a
whole bunch of model specifications and plots them. Is there an
available software package that performs this type of analysis?
I have replicated a model in which 0% of counterfactuals are in
the convex hull, and I'd like some way to visually demonstrate if/
how results change across different model specifications. Since it
looks like I cannot remove extrapolation and still have
observations left, I thought this type of analysis would
demonstrate my point. If somebody has a better suggestion, I would
be interested in that too.
Thanks,
Keith Schnakenberg
p.s. Is it still okay to use this list now that the class is over?
If not, feel free to email me privately.
____________________________
Keith Schnakenberg
MA Student
Department of Political Science
San Diego State University
keith.schnakenberg at
gmail.com
____________________________
_______________________________________________
gov2001-l mailing list
gov2001-l at
lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l at
lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l