15 Jul
2009
15 Jul
'09
9:13 a.m.
Dear Amelia users/authors:
I was able to impute CSTS data with logical bounds before the latest update
to Amelia. I was wondering whether anyone else experienced similar problems
with the latest update (maybe I am just doing something wrong...)
Thanks for any help, best,
Milan
---
Milan Svolik
Assistant Professor
Department of Political Science
University of Illinois at Urbana-Champaign
https://netfiles.uiuc.edu/msvolik/www/
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
15 Jul
15 Jul
9:42 a.m.
Hi Milan,
Are you receiving an error message from Amelia? Or are the imputation
falling outside of the specified bounds?
In addition to the bounds argument, there is another way to specify
two-sided bounds using transformations. If you can transform your
variable into a (0,1) variable (that is, a proportion between the two
bounds), then you can pass this transformed variable to the logistic
argument (lgstc), which will force Amelia to always impute it between
0 and 1. After you get the output back from Amelia, you can put it
back on to the original scale with code like this:
for (i in 1:a.out$m) {
a.out$imputations[[i]]$myvar <- scale*a.out$imputations[[i]]$myvar + offset
}
Filling in the scale and offset values according to the transformation
to a (0,1) variable. This might be of help to someone trying to
implement bounds.
In any case, please give us a bit more detail on your problem and we
can try to run it down.
Thanks,
matt.
On Wed, Jul 15, 2009 at 9:13 AM, Milan Svolik<msvolik(a)gmail.com> wrote:
Dear Amelia users/authors:
I was able to impute CSTS data with logical bounds before the latest update
to Amelia. I was wondering whether anyone else experienced similar problems
with the latest update (maybe I am just doing something wrong...)
Thanks for any help, best,
Milan
---
Milan Svolik
Assistant Professor
Department of Political Science
University of Illinois at Urbana-Champaign
https://netfiles.uiuc.edu/msvolik/www/
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
6:12 p.m.
Milan---
Recommend you make sure you have the very latest version of Amelia (v.
1.2-10)...
I also was having trouble...
The imputation would run, compute all the parameters, and then print,
"Error in matrix(NA, nrow = AMn.ss, ncol = AMp) :
non-numeric matrix extent"
...But I found out I was one iteration behind on the Amelia updates...
The newest version (1.2-10) fixed this problem (...and also runs much
faster!...)
Wayne
-----Original Message-----
From: owner-amelia_at_lists_gking_harvard_edu(a)mail.hmdc.harvard.edu
[mailto:owner-amelia_at_lists_gking_harvard_edu@mail.hmdc.harvard.edu] On
Behalf Of Matt Blackwell
Sent: Wednesday, July 15, 2009 09:42
To: Milan Svolik
Cc: amelia(a)lists.gking.harvard.edu
Subject: Re: [amelia] CSTS data with boundary priors
Hi Milan,
Are you receiving an error message from Amelia? Or are the imputation
falling outside of the specified bounds?
In addition to the bounds argument, there is another way to specify
two-sided bounds using transformations. If you can transform your
variable into a (0,1) variable (that is, a proportion between the two
bounds), then you can pass this transformed variable to the logistic
argument (lgstc), which will force Amelia to always impute it between
0 and 1. After you get the output back from Amelia, you can put it
back on to the original scale with code like this:
for (i in 1:a.out$m) {
a.out$imputations[[i]]$myvar <- scale*a.out$imputations[[i]]$myvar +
offset
}
Filling in the scale and offset values according to the transformation
to a (0,1) variable. This might be of help to someone trying to
implement bounds.
In any case, please give us a bit more detail on your problem and we
can try to run it down.
Thanks,
matt.
On Wed, Jul 15, 2009 at 9:13 AM, Milan Svolik<msvolik(a)gmail.com> wrote:
Dear Amelia users/authors:
I was able to impute CSTS data with logical bounds before the latest
update
to Amelia. I was wondering whether anyone else
experienced similar
problems
with the latest update (maybe I am just doing
something wrong...)
Thanks for any help, best,
Milan
---
Milan Svolik
Assistant Professor
Department of Political Science
University of Illinois at Urbana-Champaign
https://netfiles.uiuc.edu/msvolik/www/
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
<http://1429236.signature1.mailinfo.com/confirm2.6/00050B08/0601024E/0000094
D/0621191424.jpg>
8:55 p.m.
Hi Matt and Wayne,
Thanks for your help. I updated to the very latest version (1.2-12). But I
am still having trouble with the bounds priors. This is the message I am
getting:
Error in AMr1[, varPriors, drop = FALSE] : incorrect number of dimensions
Nonetheless, I will follow Matt's advice and work with transformed variables
instead. Without the bounds, everything seems to be working fine.
Best,
Milan
-----Original Message-----
From: Matt Blackwell [mailto:mblackwell@gmail.com]
Sent: Wednesday, July 15, 2009 9:42 AM
To: Milan Svolik
Cc: amelia(a)lists.gking.harvard.edu
Subject: Re: [amelia] CSTS data with boundary priors
Hi Milan,
Are you receiving an error message from Amelia? Or are the imputation
falling outside of the specified bounds?
In addition to the bounds argument, there is another way to specify
two-sided bounds using transformations. If you can transform your
variable into a (0,1) variable (that is, a proportion between the two
bounds), then you can pass this transformed variable to the logistic
argument (lgstc), which will force Amelia to always impute it between
0 and 1. After you get the output back from Amelia, you can put it
back on to the original scale with code like this:
for (i in 1:a.out$m) {
a.out$imputations[[i]]$myvar <- scale*a.out$imputations[[i]]$myvar +
offset
}
Filling in the scale and offset values according to the transformation
to a (0,1) variable. This might be of help to someone trying to
implement bounds.
In any case, please give us a bit more detail on your problem and we
can try to run it down.
Thanks,
matt.
On Wed, Jul 15, 2009 at 9:13 AM, Milan Svolik<msvolik(a)gmail.com> wrote:
Dear Amelia users/authors:
I was able to impute CSTS data with logical bounds before the latest
update
to Amelia. I was wondering whether anyone else
experienced similar
problems
with the latest update (maybe I am just doing
something wrong...)
Thanks for any help, best,
Milan
---
Milan Svolik
Assistant Professor
Department of Political Science
University of Illinois at Urbana-Champaign
https://netfiles.uiuc.edu/msvolik/www/
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
More info about Amelia: http://gking.harvard.edu/amelia
5420
days inactive
5421
days old
amelia@lists.gking.harvard.edu
3 comments
3 participants
participants (3)
-
Matt Blackwell -
Milan Svolik -
Wayne Thornton