Dear R listserve,

I am writing with the hope or receiving assistance using the MICE package.My 
troubles relate to post-processing imputations. I am attempting to impute 
variables for which I want to specify two conditions using post-processing 
imputation: 1) if the value of a variable representing the pesticide 
application (PESTICIDE) is coded 0 to represent that the participant did not 
conduct task then I want my duration of work in the task (DURR) and start year 
for the task variables (START_DATE) to be coded as NA; 2) If the value of the 
variable representing pesticide application task is coded 1 (did conduct task), 
then I want my duration of work in the task and start year for the task to be 
imputed  using certain boundaries (restricting the values of the imputed date 
and duration variables).

The two statements seem to work independently (post-processing statements are 
recognized); however, I can't figure out how to combine them in one post 
statement . Thus, I can't get my imputations to take the two conditions into 
account; I can either restrict the variables to be set to missing if PESTICIDE 
is 0, or I can get my imputations to be restricted to my specified boundaries, 
but I can't get R to take both conditions into account simultaneously.

The following is the code that I am using. I am highlighting the two post 
processing statements that I am having trouble with.:

ini<-mice(agrican.subb,  print=FALSE, max=0)
pred<-ini$pred

pred["START_DATE",]<-1
pred["END_DATE",]<-0
pred[,"END_DATE"]<-0
pred[,"START_DATE"]<-0
pred[,"YEAR_BIRTH"]<-0
pred[,"year_ten"]<-0

pred[,"DIFF"]<-0
pred[,"ID"]<-0
pred[,"CORN"]<-0

method<-c(""," ", "", "", "", "norm", "", "pmm", "", "", "", "", "", "", "norm")

post<-ini$post

post["START_DATE"]<-"imp[[j]][,i]<-squeeze(imp[[j]][,i],c(agrican.subb$YEAR_TEN,
 agrican.subb$ENROLLMENT_YEAR))"
post["START_DATE"]<-"imp[[j]][p$data$PESTICIDE[!r[,j]]<1,i]<-NA"

post["DIFF"]<-"imp[[j]][,i]<-squeeze(imp[[j]][,i],c(0, 0))"
#post["DIFF"]<-"imp[[j]][p$data$PESTICIDE[!r[,j]]<1,i]<-NA"

imp<-mice(subb,post=post, pred=pred, seed=2332, method=method, maxit=3)

I will appreciate any advise that you can offer.

Thanks in advance!
Leah

Postdoctoral fellow
Section of Environment and Radiation
International Agency for Research on Cancer
150, cours Albert Thomas
69372 Lyon Cedex 08
France
schina...@fellows.iarc.fr
Tel: [+33] (0)472 73 84 85


-----------------------------------------------------------------------
This message and its attachments are strictly confidenti...{{dropped:11}}

______________________________________________
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