Dear list members,
my problem is to analyze multiple variables by using a simple loop.
Without a loop, no problem:
fit1 <- fit.mult.impute(variable_from_MIC_datset ~ group, fitter=ols,
xtrans=imp.dmat)
Without a multiple imputation data set, that works:
vars <- c(
"lverb.ona", "l2", "lalles.ona", "vl", "notdmer.ona", "pisalern.ona",
"lern",
"pswa", "pswn", "pska", "pskd", "pskt", "pkoo", "pkom", "isall", "vorwisse",
"im", "ke", "sb", "an", "akog", "aemo", "amb", "stru")
for(i in vars)
{ lm(dmat[,i]~ group) }
However, I do not know how to specifiy the following correctly with the
MICE object:
for(i in vars)
{
fit1 <- fit.mult.impute(?????? ~ group, fitter=ols, xtrans=imp.dmat)
}
Looking into the structure of MICE objects, each of the x multiple
imputed datasets is stored in a data.frame within a list. That list is
named after the respective variable.
If I specify just the variable like
imp.dmat$imp$var_name
that does not work, because it is just the name of the list element.
Otherwise by specifying one of the columns with the list element, the
multiple datasets would be missed.
I thought about using eval() or expr() but without real succes.
Every hint is appreciated,
best regards
leo gürtler
______________________________________________
R-help@stat.math.ethz.ch mailing list
https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide! http://www.R-project.org/posting-guide.html
