Dear all,
how do you calculate summary statistics and other quantities of interest or plots from variables in my m=5 imputed data sets after running Amelia?
I need:
- summary statistics table (min,max,mean)
- bivariate correlation matrix between variables
- time series plots for some variables
Has anyone done this and could you let me know how you went about it? I hesitate to merge all 5 imputed data sets into one for this. What is the standard? Do you do this only with the original data, but not after imputation?
Please help!
Best,
Nicole
Nicole Janz, PhD Cand.
Lecturer at Social Sciences Research Methods Centre 2012/13
University of Cambridge
Department of Politics and International Studies
www.nicolejanz.de | nj248(a)cam.ac.uk | Mobile: +44 (0) 7905 70 1 69 4
Skype: nicole.janz
Hello,
I am using Amelia II and have a couple of queries.
(1) I would like to ensure the reproducibility of the imputations so I
include 'seed=2345' in the Amelia line command. However, I notice
that, if I repeat the run with exactly the same line command, the
chain lengths differ at each imputation. Will I be getting the same
imputations?
(2) At the end of an imputation (with no parameters significantly
changed since the last iteration), I sometimes get the message:
error: chol(): failed to converge
together with a little window headed 'Microsoft Visual C++ Runtime
library' and the message ' This application has requested the Runtime
to terminate it in an unusual way. Please contact the application's
support team for more information.'
How should I interpret and respond to this?
Thanks.
Steve
Dear all,
I have m=5 imputed data sets with a time-series cross-country structure, and I keep them in a list structure (they are not an Amelia object). I would like to calculate:
- summary statistics
- bivariate correlation matrix between variables
- AIC (after running Amelia)
- time series plots for each country and my dependent variable
Has anyone done this and could you let me know how you went about it?
With AIC, I'm thinking of extracting the AIC of each model (run on the 5 data sets), and then just taking the mean. About the summary statistics and correlation matrix I'm not sure how to proceed to maintain the within and between variance of my m=5 imputed data sets. I'm also not sure how to show a time series plot if I have 5 data sets.
Best,
Nicole
Nicole Janz, PhD Cand.
Lecturer at Social Sciences Research Methods Centre 2012/13
University of Cambridge
Department of Politics and International Studies
www.nicolejanz.de | nj248(a)cam.ac.uk | Mobile: +44 (0) 7905 70 1 69 4
Skype: nicole.janz
Hello,
I am new to R and Amelia. I have managed to pull a small Stata
dataset into Amelia, generate the imputations, and then push the
resulting imputed datasets back into Stata for analysis using the
-xtreg- command. So far, so good.
My actual dataset comes from the World Bank. It consists of about 50
variables for 40 countries over a 20 year period. There is a lot of
missingness. There is also a lot of non-normality. However, the
Amelia program guide 1.7.1 suggests that non-normality does not
necessarily mean problems with the imputation model, so I have started
with the variables untransformed.
However, the EM algorithm is generating a lot of error messages
'error: inv(): matrix appears to be singular'. I am not sure how best
to respond to this.
(1) Should I be looking to drop highly correlated variables in the
model? But how high is 'high'? Is there an easy way to identify the
'problem' variable(s)?
(2) Should I be looking to drop variables with a high degree of missingness?
(3) Is the lack of invertibility related to non-normality?
If all three are relevant, which should I prioritize?
Any suggestions/guidance would be most welcome.
Thanks.
Steve
Hello,
I'm trying to impute a dataset with a variable to indicate cross-sections. When I impute each geographic unit on its own or in aggregate without the cross-section specified, the imputation completes without error. However, when I impute the dataset with the argument cs = "region", I receive the following error:
Error in yy %*% unique(na.omit(x.orig[, i])) : non-conformable arguments
My code looks as follows:
imputed <- amelia(W, m = 1, cs = "region",
idvars = c("bunch", "of", "variables"),
noms = c("bunch", "more", "variables"))
Any help would be greatly appreciated.
Thanks,
Gregory
