For an example along the lines that Gary suggests, you can see the
following article by David Stasavage and I:
It doesn't implement the method in Gelman and King (1998), but you can do
things similar. I'm unsure how you're defining pooled data. If you mean
time-series-cross-sectional data, then it was designed for that, in addition
to simpler problems. if you mean that you have (say) two cross-sectional
surveys with some of the same questions but some missing from one of the
surveys, then you can stack up the surveys and use Amelia.
Gary
On Sat, 13 Jan 2007, Christopher Parker wrote:
Hi all,
Two questions. First, can Amelia be used to impute questions that are not
observed in dataset "A" from dataset "B" e.g., Gelman and King
(1998)?
Second, if so, is it appropriate to do so with pooled data. Is it
possible, for instance, to impute the altruism questions from the 2002 GSS
into pooled ANES data that combined the 2002 with 2004 ANES?
Thanks in advance for your advice.
cp
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive:
http://lists.gking.harvard.edu/?info=amelia
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: