26 Apr
2011
26 Apr
'11
2:41 p.m.
I think the only package that will do this is cem(). not cem within
matchit. (its not that well tested, but it should work if you recode the
continuous treatment into categories first)
Gary
--
*Gary King* - Albert J. Weatherhead III University Professor - Director,
IQSS - Harvard University
GKing.Harvard.edu <http://gking.harvard.edu/> - King(a)Harvard.edu -
@kinggary<http://twitter.com/kinggary>- 617-500-7570 - Asst 495-9271 -
Fax 812-8581
On Tue, Apr 26, 2011 at 2:35 PM, Jackelyn Hwang <jihwang(a)fas.harvard.edu>wrote;wrote:
Hi class,
Our group is trying to do matching with a continuous treatment variable.
We've tried to use MatchIt with cem and cem() itself but have failed to get
matches. We've also tried to coarsen the treatment into categories, but the
methods are still not working. We would ideally like to match groups that
are at least two category levels apart out of 4 categories (i.e., 1 and 3; 2
and 4). If anyone knows anything about multicategory treatments for cem,
which the documentation says it can do, please email me!
Thanks,
Jackie
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l
_______________________________________________
gov2001-l mailing list
gov2001-l(a)lists.fas.harvard.edu
http://lists.fas.harvard.edu/mailman/listinfo/gov2001-l