Hi Nathan,
We changed some of the syntax of Amelia in recent versions to more
closely align it with the rest of R. I imagine that this could problem
could be a result of this change. Try giving this a shot and see what
happens: change polytime = 0 to polytime =NULL. It will produce the
same imputations, but there is a slight problem when both intercs and
polytime are 0.
Let us know if that doesn't work.
Cheers,
matt.
On Thu, Nov 12, 2009 at 6:35 PM, Nathan Paxton <napaxton(a)fas.harvard.edu> wrote:
I am using Amelia 1.2-12 and R 2.9.1.
I am using a batch file for R that I haven't used in a year, and as I last
recall, it worked without generating errors. But as I ran it this time, I
got the following error. This occurred in imputation 1, "run" 1000-1200 (I
tried a few times)....
Error in dimnames(impdata$covMatrices)[[1]] <- prepped$theta.names :
length of 'dimnames' [1] not equal to array extent
In addition: Warning message:
In amcheck(x = x, m = m, idvars = numopts$idvars, priors = priors, :
You've set the polynomials of time to zero with no interaction with
the cross-sectional variable. This has no effect on the imputation.
Although I've gotten decent with Amelia, I'm not quite sure what this means.
Best,
-Nathan
----------
Nathan A. Paxton, Ph.D.
Dept. of Government, Harvard University
napaxton AT fas DOT harvard DOT edu
http://www.fas.harvard.edu/~napaxton
========================================================
If every professor who backed a lunatic politician were to be sacked, half
the interesting minds in academia would be lost.
- The Economist, 5 Jan 2002
A morning without coffee is like something without something else.
========================================================
-
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