That's right!
Your advice is in the right direction and with little adjustments it did
the job. However, I admit it was tricky and the result looks a bit
artisanal and needs some polishing that I will do by hand in the tex code.
Is it possible that there is no way to get nicely latex formatted tables
concerning multiply imputed data-set?
But maybe I should open a separate thread on this.
Thanks a lot for your kind and patient help.
Best regards,
f.


On 18 August 2013 15:56, Christopher Desjardins <cddesjard...@gmail.com>wrote:

> Seems you're after the pooled results. Would the following work?
>
>
>
> library(Amelia)
> library(Zelig)
> library(xtable)
>
> data(africa)
>
> m = 10
> imp1 <- amelia(x = africa,cs="country",m=m)
> imp2 <- amelia(x = africa,cs="country",m=m)
> lm.imputed1 <- zelig(gdp_pc ~ trade + civlib, model="ls",data = imp1)
> lm.imputed2 <- zelig(gdp_pc ~ trade + civlib, model="ls",data = imp2)
>
> lm1 <- as.data.frame(summary(lm.imputed1)$coef)
> lm2 <- as.data.frame(summary(lm.imputed2)$coef)
> lm1[,2] <- ifelse(lm1[,4]<.001,paste(lm1[,2],"***",sep=" "),
>                   ifelse(lm1[,4]<.01,paste(lm1[,2],"**",sep=" "),
>                          ifelse(lm1[,4]<.05,paste(lm1[,2],"*",sep=" "),
>                                 ifelse(lm1[,4]<.1,paste(lm1[,2],".",sep="
> "),lm1[,2]))))
>
> lm2[,2] <- ifelse(lm2[,4]<.001,paste(lm2[,2],"***",sep=" "),
>                   ifelse(lm2[,4]<.01,paste(lm2[,2],"**",sep=" "),
>                          ifelse(lm2[,4]<.05,paste(lm2[,2],"*",sep=" "),
>                                 ifelse(lm2[,4]<.1,paste(lm2[,2],".",sep="
> "),lm2[,2]))))
>
> ## ONE OPTIONS ##
> lms <- as.data.frame(cbind(lm1[,1],lm2[,1],lm1[,2],lm2[,2]))
> rownames(lms) <- rownames(lm1)
> colnames(lms) <- c("Imp1.Est","Imp2.Est","Imp1.SE","Imp2.SE")
> xtable(lms)
>
> ## OR ##
> xtable(cbind(lm1[,1:2],lm2[,1:2]))
>



-- 
Francesco Sarracino, Ph.D.
https://sites.google.com/site/fsarracino/

        [[alternative HTML version deleted]]

______________________________________________
R-help@r-project.org mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.

Reply via email to