Hello all,
I am having a bit of trouble using the save session feature on Amelia. I am currently running Amelia II on a Windows XP machine. After loading my data and entering in the basic options I need to impute my data set (identifying the cross-section and time-series variables), I went ahead and began entering my 150 analytical priors. Once done, I saved the session (naming the file the same thing as my dataset as the manual indicates), and then had to leave to deal with another matter. I returned later and attempted to load my previous session. When I did this, I ended up with the dataset correctly loaded and my time-series and cross-sectional data identified. Unfortunately, all of my priors were gone.
I subsequently tried the same process two more times (although I entered in only 3 priors) in order to see if I had made an obvious error in the process of saving. The results were sadly the same. I also attempted saving twice, once before beginning to enter my priors and once afterwards, to see if that could be the problem. Again, no results.
I'm sure I'm making an error somewhere, but I'm not sure what it could be.
Thanks,
-IM
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
Hallo Gary,
And al the others of the list.
First of al, thanks for the fast replay.
My problem in short: when I mark a polytomous nominal variable (party
affiliation)in my dataset as "nominal" in Amelia view and I want to
overimpute it Amelia tells me that "The variable you selected doesn't exist
in the Amelia output because it wasn't imputed". However, when I select
"compare" Amelia shows me the density of the observed and the imputed values
as well.
What Im using:
R: Version 2.4.0
Amelia: latest Version
What i have done so fare:
I am using a dataset from Iversen and Soskice (2001) which is available
online at:
http://www.people.fas.harvard.edu/~iversen/data/preferences.htm => Dataset 1
I dropt the variables v23, v26, v30, v31 which are combined to the index "P"
in the dataset. The same I have done with variables v25 v32, which are
combined to the index "Postmat".
The resulting dataset I have loaded in AmeliaView. Then I have selected
"Nominal" for variable v223 in the "Variables" Dialog box which is the party
affiliation variable I mentioned. When I start the imputation, it runs quite
well.
Best wishes,
Tim
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
Dear all,
I earlier posted a message about what struck me as something odd about
how AmeliaView is treating range priors. Subsequently I've tried
imputing the same data set (using the same parameters) in R itself.
(I'm running the most recent versions of both R and Amelia on a
Windows XP system.) I've tried many "tweaks" to the entire-variable
range priors (exactly which variables receive a range prior; the
minimum and maximum values; the confidence coefficient), and no matter
what I do, whenever I include range priors, Amelia returns an error of
the following sort:
Error in chol(copy.theta[c(FALSE, m[ss, ]), c(FALSE, m[ss, ])]) :
the leading minor of order 3 is not positive definite
Amelia runs without a hitch for this data set once I eliminate the
range priors.
Any help would be much appreciated.
Best,
Heather Stoll
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
I suggest you update to the lastest version of Amelia and try again.
that should do it. If not, tell us more details about what you did.
Gary
On Wed, 7 Mar 2007, Tim Müllenborn wrote:
> Hallo out there,
>
> I'm facing the same problem as saeid has mentioned in the list at the end of
> februar.
>
> When i select a variable which i have treated as nominal (it is party
> affiliation variable on a left-right scale) to be overimputed amelia returns
> "The variable you selected doesn't exist in the amelia output because it
> wasn't imputed"
>
> It would be nice when anybody can give me a hint,
>
> Thanks
> Tim
>
>
Hallo out there,
I'm facing the same problem as saeid has mentioned in the list at the end of
februar.
When i select a variable which i have treated as nominal (it is party
affiliation variable on a left-right scale) to be overimputed amelia returns
"The variable you selected doesn't exist in the amelia output because it
wasn't imputed"
It would be nice when anybody can give me a hint,
Thanks
Tim
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
Dear List,
In my dataset of 28 variables I have two that are multinomial distributed (religious denomination and main work activity). When I set them to nominal in the variables option in the very newest version of Amelia, I get the following error code:
There was an unexpected error in the execution of Amelia.
Double check all inputs for errors and take note of the error message:
Error in chol(theta[c(FALSE, m[ss, ]), c(FALSE, m[ss, ])]) :
the leading minor of order 1 is not positive definite
Setting the two nominal variables to "no transformation", Amelia runs. Surely there is a better solution?
Thank you very much for your help!
All the best,
Katharina
--
Katharina Grünberg
Department of Mathematics and Statistics
Lancaster University
Lancaster
UK
-
Amelia mailing list served by Harvard-MIT Data Center
[Un]Subscribe/View Archive: http://lists.gking.harvard.edu/?info=amelia
