Hi David,
This is a bug with the "ords" command that we (finally) tracked down,
thanks to your test case. The newest version of Amelia (1.2-14) fixed
this problem.
If you have R 2.10.x, you can update your version of Amelia with the
following command:
install.packages("Amelia", repos="http://gking.harvard.edu")
Otherwise, the updated version will be on CRAN whenever they can
process it, probably in the next few days.
I hope this helps and please let us know if you have any further issues.
Thanks!
matt.
On Mon, Nov 16, 2009 at 11:28 AM, David Hugh-Jones
<davidhughjones(a)gmail.com> wrote:
Hi all,
I get a strange bug when using the "disperse" function with some
ordinal variables. The output of Amelia correctly gives these ordinal
variables integer values. However, they are stored in R as numeric
types. When I call "disperse", it calls the Amelia internal function
"amcheck". A few of the ordinal variables then fail the test
if (any(unique(na.omit(x[, i]))%%1 != 0))
and the disperse function fails.
Here's a simple test case that fails for me:
tmp <- data.frame(a=1:10000, b=round(rnorm(100000,0,3)))
tmp$b[1:2000*5] <- NA
a.out.try <- amelia(tmp, ords="b")
disperse(a.out.try)
David Hugh-Jones
Post-doctoral Researcher
Max Planck Institute of Economics, Jena
http://davidhughjones.googlepages.com
-
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