14 Feb
2012
14 Feb
'12
10:04 a.m.
Hi Prashant,
Perhaps you could create a categorical variable that is numeric instead of character-based
and use that instead. This might get around some of the problems with imb. (Character
encoding can be quite tricky.)
Cheers,
matt.
On Tuesday, February 14, 2012 at 9:32 AM, Prashant wrote:
I'm sorry for sending another e-mail -- one
correction to my previous
post is that the hometown01 variable is categorical -- and the
categories are in Chinese characters -- perhaps this is a reason why
imb is having difficulty.
thank you again,
Prashant
On Tue, Feb 14, 2012 at 10:29 PM, Prashant <presearchwork(a)gmail.com
(mailto:presearchwork@gmail.com)> wrote:
Hi Matt,
Thank you kindly for the quick reply -- I later tried with different
combinations of variables and found that the imb command works after
taking out the binary variable "hometown01". The cem command works
even when it (or any combination of variables) is included. Does it
have to do with the ~15% missing values in hometown01 perhaps?
Thank you for all of your help!
Prashant
On Tue, Feb 14, 2012 at 10:18 PM, Matt Blackwell
<blackwel(a)fas.harvard.edu (mailto:blackwel@fas.harvard.edu)> wrote:
Hi Prashant,
Have you tried running the imb function on a smaller set of the variables?
The `cem` command calls imbalance at some point, so this should work.
Another quick potential solution is to reinstall the CEM package and see if
that fixes the issue.
Hope that helps!
Cheers,
matt.
~~~~~~~~~~~
Matthew Blackwell
PhD Candidate
Institute for Quantitative Social Science
Department of Government
Harvard University
url: http://www.mattblackwell.org
On Tuesday, February 14, 2012 at 5:59 AM, Prashant wrote:
Hi,
I'm get an error message when I run imb in Stata (cem runs fine):
L1meas(): 3301 subscript invalid
imbalance(): - function returned error
<istmt>: - function returned error
r(3301);
My code is something like:
imb new_age borninbeijing female hometown01 BJ_hukou Rural_hukoutype
fatheredu motheredu ifpreschooling schoolingathome,
treatment(public_yn)
Any help would be most appreciated!
Thanks,
Prashant
-
--
cem Mailing List, served by HUIT
Send messages: cem(a)lists.gking.harvard.edu (mailto:cem@lists.gking.harvard.edu)
[un]subscribe Options: http://lists.gking.harvard.edu/?info=cem
More information on cem: http://gking.harvard.edu/cem
Cem mailing list
Cem(a)lists.gking.harvard.edu (mailto:Cem@lists.gking.harvard.edu)
https://lists.gking.harvard.edu/mailman/listinfo/cem