[R] computationally singular error with mice()

[Weidong Gu]

Hi Fei,

I wouldn't worry to much about categorical variables for mice. Mice
would use logisitic regression for binary and polytomous logistic
regression for categorical variables with >2 levels. However, you
should not include factors with a lot of levels, saying>30, in
imputation models because it would require a lot of dummy variables.

Another thing is that not excluding variables you would use in
substantive analysis. Otherwise, estimation would be biased.

Weidong



On Sat, Nov 26, 2011 at 12:07 PM, Fei <fayechen0807 at 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
>
> 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.
>