computationally singular error with mice()

Hi Fei,

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 at 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/