If I understand your question and if I understood Dr Harrell's
description of the Hmisc/Design packages capabilities, this should all
be available. I have been blessed with a dataset so large and with so
relatively few missing values that I have not yet needed to put his
methods into practice, but you should consider his function transcan
in Hmisc.
On Mar 12, 2009, at 8:44 AM, Elsaesser, Chris wrote:
> I need a package that can compute missing values of n-dimensional
> vectors for n > 2. This is a kind of interpolation, complicated in
> dimensions higher than 2. The idea is that I have a set of fully
> specified vectors (i.e., with no missing values) and I get a new
> vector that has one or more missing attributes; I need to fill in
> the missing values with values that are, based on the 'training
> set', representative.
>
> EMV appeared to be the solution but it has been taken off CRAN and I
> can't seem to install the archived version by hand on my Windoz
> machine.
>
> Does anyone know of an alternative to EMV?
>
> Thanks,
>
> Chris Elsaesser, PhD 703.371.7301 (m)
> Principal Engineer 703.983.4116 (o)
> Information Discovery & Understanding
> The MITRE Corporation
> McLean, VA
>
> ______________________________________________
> 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.
David Winsemius, MD
Heritage Laboratories
West Hartford, CT