From: Kosuke Imai &lt;kimai(a)Princeton.Edu&gt; Sent: Jan 18, 2010 4:01 PM To: Peter Flom &lt;peterflomconsulting(a)mindspring.com&gt; Cc: zelig(a)lists.gking.harvard.edu Subject: Re: [zelig] summary with imputations bug ... any workarounds? The point estimates can be obtained by running Zelig for each imputed model and calculating the mean of all five estimates for each parameter. Standard errors are a little tricky: see Little and Rubin or Schafer's books on multiple imputation for the formulae. If you know how to do some R programming, you can combine all simulation draws from each of the five runs. And then calculate whatever the quantities you like. Good luck, Kosuke -- Department of Politics Princeton University http://imai.princeton.edu On Mon, 18 Jan 2010, Peter Flom wrote: > OK, thanks.... any suggestions in the meantime? > > I can use SAS, but I don't like their imputation methods in this case. > > Or maybe I can just look at the five outputs, and see if any were much different from the results with casewise deletion? > > Thanks > > Peter > > > -----Original Message----- >> From: Kosuke Imai &lt;kimai(a)Princeton.Edu&gt; >> Sent: Jan 18, 2010 3:13 PM >> To: Peter Flom &lt;peterflomconsulting(a)mindspring.com&gt; >> Cc: zelig(a)lists.gking.harvard.edu >> Subject: Re: [zelig] summary with imputations bug ... any workarounds? >> >> Unfortunately, mixed effects models are not currently compartible with >> multiply imputed data sets. It's on our to-do list... >> >> Kosuke >> >> -- >> Department of Politics >> Princeton University >> http://imai.princeton.edu >> >> On Fri, 15 Jan 2010, Peter Flom wrote: >> >>> Good morning >>> >>> If I run >>> >>> <<< >>> susan.lsmixed.out <- zelig(formula = unprot_vag_sex ~ married + age + TREATMENT.ARM*time + highest_grade + income + tag(1|id), >>> data = susanMI.out$imputations, model = "ls.mixed") >>> >>> >>> summary(susan.lsmixed.out) >>>>>>> >>> >>> I get an error >>> >>> Error in x$coef : $ operator is invalid for atomic vectors >>> >>> Searching the archives, I see that others have had similar problems. Is there a workaround? >>> >>> summary(susan.lsmixed.out[[1]]) >>> >>> works fine; should I then average across the five imputed data sets? >>> >>> thanks! >>> >>> Peter >>> >>> Peter L. Flom, PhD >>> Statistical Consultant >>> Website: http://www DOT statisticalanalysisconsulting DOT com/ >>> Writing; http://www.associatedcontent.com/user/582880/peter_flom.html >>> Twitter: @peterflom >>> >>> - >>> Zelig Mailing List, served by Harvard-MIT Data Center >>> Send messages: zelig(a)lists.gking.harvard.edu >>> [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig >>> Zelig program information: http://gking.harvard.edu/zelig/ >>> >> >> - >> Zelig Mailing List, served by Harvard-MIT Data Center >> Send messages: zelig(a)lists.gking.harvard.edu >> [un]subscribe Options: http://lists.gking.harvard.edu/?info=zelig >> Zelig program information: http://gking.harvard.edu/zelig/ > > > Peter L. Flom, PhD > Statistical Consultant > Website: http://www DOT statisticalanalysisconsulting DOT com/ > Writing; http://www.associatedcontent.com/user/582880/peter_flom.html > Twitter: @peterflom >