[R] Announce: multilm
hothorn at statistik.uni-dortmund.de
Sat Sep 25 15:24:32 CEST 1999
On 24 Sep 1999, Peter Dalgaard BSA wrote:
> Torsten Hothorn <hothorn at statistik.uni-dortmund.de> writes:
> > I'm not sure if those basic methods fit most users needs, maybe wrappers for
> > special models (MANOVA) are required. I'm very grateful for any hint and
> > discussion, mails of common interest please to the R-list, others to my
> > email address.
> Now I didn't actually try this (yet), but one thing that reduces the
> practical value of these models is missing values. Is there any
> provision to handle these? Someone mentioned code to handle
> multivariate normals with missings earlier today, how do the two
> packages get along?
Laeuter and Kropf discuss the handling of missing values in their paper
Kropf et.al: Stabilized Multivariate Tests - The Inclusing of Missing
Values, Biometric Journal 39, pp. 149-169
They use the variable mean instead of NA's.
PLEASE note: there is an mistake in multilm_0.1.tar.gz, I corrected it and
> I did some special cases of this for a project a couple of years back
> where I did polynomial curve fits to repeated measurements of bone
> metabolism data. In those data, an overall rate of 10% missing data
> would result in losing half the data set for complete-case analyses.
> O__ ---- Peter Dalgaard Blegdamsvej 3
> c/ /'_ --- Dept. of Biostatistics 2200 Cph. N
> (*) \(*) -- University of Copenhagen Denmark Ph: (+45) 35327918
> ~~~~~~~~~~ - (p.dalgaard at biostat.ku.dk) FAX: (+45) 35327907
r-help mailing list -- Read http://www.ci.tuwien.ac.at/~hornik/R/R-FAQ.html
Send "info", "help", or "[un]subscribe"
(in the "body", not the subject !) To: r-help-request at stat.math.ethz.ch
More information about the R-help