Dear all,
I am running into some problems with the newest version of Amelia II and
I cannot figure out exactly why, and I hope you can shed some lights on
these issues that I am running into.
I have successfully imputed 10 dataset by using Amelia II in R. The
imputed datasets look good, so is the overdispersed graph. However, when
I try to run "overimpute" and "compare.density" to conduct more
diagnostics for some specific variables, I am getting the following
error messages:
> compare.density(a.out, var =13)
Error in density.default(vars[!is.na(vars)], na.rm = TRUE) :
argument 'x' must be numeric
> overimpute(a.out, var =13)
Error in segments(xplot[ci.order], lowers[ci.order], xplot[ci.order], :
invalid first argument
I double-check and make sure variable 13 is indeed numeric, but they
have non-English labels.
I have used an older version of Amelia II and imputed a similar dataset
a couple years before, and I have no problem using these diagnostics
commands in R. Now that I have the most up-to-date version of Amelia II
on a new computer, I am running into these problems.
Any suggestions will be greatly appreciated. Thanks a lot in advance.
Best,
X
--
Xiaobo Lü
Assistant Professor
Bush School of Government and Public Service
Texas A&M University
Tel: (979) 845-6510
Fax: (979) 845-4155
Email: xlu(a)bushschool.tamu.edu
Website: http://www.xiaobolu.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
Hello,
I'm wondering if someone could provide a little bit of information about how observation-level priors work for a variable that is also being transformed?
I'll apply my question to the running example found in the AMELIA documentation (Honaker, King, and Blackwell, 2011). Here, using the option logs="tariff" seems appropriate as discussed on pages 18 and 19. Imagine now that I also want to include observational-level priors about tariff, as discussed on pages 24-26. Will my prior also be converted into log-space? I assume without the transformation option, my prior is assumed to be normally distributed, based on the confidence interval supplied by my five column priors matrix. What is the distribution now that I've also specified the log transformation on the tariff? To avoid this, should I simply not transform tariff?
Thanks for any help that anybody can offer.
Joe Dieleman
University of Washington
Institute for Health Metrics and Evaluation