Hi Andrew,
Looking at the webpage, it is more extensive than Amelia in some ways.
Amelia is an approximation to MVN imputation only. Looks like
Stata's mi has that but also includes various options for single
variable imputation as well as chained equations for monotone
missingness. Add in that mi has a nice estimation feature, and what
you are getting is basically an amalgam of what exists for Amelia,
MICE, and Zelig for R. Now, if you are working with data that have
missing data on multiple values and the pattern is not monotone---this
is a common scenario----then the imputation step does exactly the same
thing as Amelia (MVN imputation). Seems like you could also import
data from Amelia and use mi's features to do estimation (regressions,
etc.) on the imputed datasets. So, under the common "non-monotone
missingness" scenario, it's just Amelia by another name.
Cyrus
On Wed, Sep 23, 2009 at 2:13 PM, Andrew Therriault <[log in to unmask]> wrote:
> Has anyone worked with the new MI feature in Stata 11? :
>
> http://www.stata.com/stata11/mi.html
>
> From a convenience standpoint it sounds great, but I wonder how it stacks up
> methodologically to Amelia and other existing tools. Also, Stata's MI
> interface appears to be highly customizable with regard to specific methods
> for imputing missing data. This is great for those who are highly-versed in
> such techniques, but might be daunting to the only-moderately familiar. Can
> anybody suggest their preferred settings, or recommend a straightforward
> resource for choosing among the options?
>
> Thanks,
>
> Andrew Therriault
>
> Doctoral Candidate
> Department of Politics
> New York University
> [log in to unmask]
>
> **********************************************************
> Political Methodology E-Mail List
> Editors: Xun Pang <[log in to unmask]>
> Jon C. Rogowski <[log in to unmask]>
> **********************************************************
> Send messages to [log in to unmask]
> To join the list, cancel your subscription, or modify
> your subscription settings visit:
>
> http://polmeth.wustl.edu/polmeth.php
>
> **********************************************************
>
--
Cyrus Samii
Political Science
Columbia University
[log in to unmask]
Burundi Survey: www.columbia.edu/~cds81/burundisurvey/
ISERP Statistical Consulting:
www.iserp.columbia.edu/services/statistical_consulting.html
Comparative Political Economy Blog: cpecolumbia.blogspot.com
**********************************************************
Political Methodology E-Mail List
Editors: Xun Pang <[log in to unmask]>
Jon C. Rogowski <[log in to unmask]>
**********************************************************
Send messages to [log in to unmask]
To join the list, cancel your subscription, or modify
your subscription settings visit:
http://polmeth.wustl.edu/polmeth.php
**********************************************************
|