[R] Experimental Design with R

S Ellison S.Ellison at lgc.co.uk
Tue Aug 28 19:33:20 CEST 2007

A full factorial can be generated using expand.grid.
Try AlgDesign, BHH2 and conf.design for various fractional and D-optimal design generators; they all work.
I haven't tried the experiment package yet, though the synopsis looks interesting.
The 'Design' package itself I found disappointing for the kind of industrial experiments that Taguchi designs are aimed at; that clearly has other (and rather bigger!) aims; a case of differing interpretations of what 'Design' really is.

AFAIK, though, none of those do Taguchi designs, response surface designs or much on mixture designs as such... and there seem to be no R tools targeted at process/formulation optimisation at the analysis stage. For example, there isn't a simple-to-use quadratic response surface modelling widget that allows you to drop terms quickly and easily and then find maxima on the resulting fitted surface (assuming you trust it as far as you can spit). It can all be done with lm, optim and a bit of patience, but not exactly intuitively.  Bit of a gap in the R arsenal, that, I felt. Sadly (before someone points out the obvious remedy) not one I have time to write a package for myself. 

But I too, may have missed something and would welcome correction.

Steve Ellison

>>> Marc BERVEILLER <marc.berveiller at edf.fr> 28/08/2007 14:11:53 >>>
Dear R-users,

I want to know if there is a package that allows to define different experimental designs (factorial, orthogonal, taguchi)
and to compare them.
I don't found one in the R-web site, but it is possible I missed it!

Thank you in advance


R-help at stat.math.ethz.ch mailing list
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html 
and provide commented, minimal, self-contained, reproducible code.

This email and any attachments are confidential. Any use, co...{{dropped}}

More information about the R-help mailing list