Hi Fei, On Sat, Nov 26, 2011 at 9:07 AM, Fei <fayechen0...@hotmail.com> wrote: > Hi Josh, > > Thanks for the kind reminder of posting the dataframe on. My dataframe > contains lots of categorical variables, which seems to be problematic. For > instance, > > dob status edu mrext > 1111 married highschool yes, full time
Still not exactly a useable dataset, but here is a snippet of code I used: ############################################################################## # Multiple Imputation Model # ############################################################################## ## specify the predictor matrix for the imputation pred.matrix <- rbind( VFQRoleDifficulties1 = c(0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1), MOODVision1 = c(1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1), MOODImpact1 = c(1, 1, 0, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 1), [snip] SocialFunctioning1 = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 0, 1, 1), RoleEmotional1 = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 0, 1), MentalHealth1 = c(1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 1, 0, 1, 1, 1, 1, 1, 0, 0, 1, 1, 1, 1, 1, 1, 1, 0)) ## set rownames to column names of the data (this is a square matrix) colnames(pred.matrix) <- colnames(dat) ## Set the methods used to impute each variable imp.method <- c( VFQRoleDifficulties1 = "pmm", MOODVision1 = "pmm", MOODImpact1 = "pmm", [snip] SocialFunctioning1 = "pmm", RoleEmotional1 = "pmm", MentalHealth1 = "pmm" ) ## Create multiply imputed dataset datimp <- mice(data = dat, m = 500, method = imp.method, predictorMatrix = pred.matrix, seed = 1, print = FALSE) Basically you can write a k x k matrix where k is the number of variables in your dataset. This can control what variables are used in the imputation model for each variable (all 0s would mean no variables). You can also pass a k length character vector controlling the method used for each variable. You can also control the order mice goes in. Cheers, Josh > > Do you know how to specify the imputation methods and the visitSquence so > that those categorical variables are not involved in the imputation process? > Thank you. > > Fei > > > > -- > View this message in context: > http://r.789695.n4.nabble.com/computationally-singular-error-with-mice-tp4109583p4110776.html > Sent from the R help mailing list archive at Nabble.com. > > ______________________________________________ > 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. > -- Joshua Wiley Ph.D. Student, Health Psychology Programmer Analyst II, ATS Statistical Consulting Group University of California, Los Angeles https://joshuawiley.com/ ______________________________________________ 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.
