Skip to content

Multiple imputation

3 messages · Jonck van der Kogel, Frank E Harrell Jr, John Fox

#
Hi all,
I'm currently working with a dataset that has quite a few missing 
values and after some investigation I figured that multiple imputation 
is probably the best solution to handle the missing data in my case. I 
found several references to functions in S-Plus that perform multiple 
imputation (NORM, CAT, MIX, PAN). Does R have corresponding functions?
I searched the archives but was not able to find anything conclusive 
there.
Any help on this subject is much appreciated.
Thanks, Jonck
#
On Thu, 12 Jun 2003 23:57:45 +0200
Jonck van der Kogel <jonck at vanderkogel.net> wrote:

            
Look at the aregImpute function in the Hmisc package (http://hesweb1.med.virginia.edu/biostat/s/Hmisc.html).  aregImpute uses the bootstrap, predictive mean matching, and flexible additive regression models to do multiple imputation.  In one simulation study it performs as well as MICE but it runs much faster and does not assume linearity in the imputation models.  I hope that someday we'll have simulation studies comparing aregImpute with NORM.
---
Frank E Harrell Jr              Prof. of Biostatistics & Statistics
Div. of Biostatistics & Epidem. Dept. of Health Evaluation Sciences
U. Virginia School of Medicine  http://hesweb1.med.virginia.edu/biostat
#
Dear Jonck,

In addition, there are ports of both norm and mix in the 
contributed-packages section of CRAN.

Regards,
  John
At 07:48 PM 6/12/2003 -0400, Frank E Harrell Jr wrote:
-----------------------------------------------------
John Fox
Department of Sociology
McMaster University
Hamilton, Ontario, Canada L8S 4M4
email: jfox at mcmaster.ca
phone: 905-525-9140x23604
web: www.socsci.mcmaster.ca/jfox