Are you familiar with :
@article{box,:hunt:1962,
Author = {Box, G. E. P. and Hunter, William G.},
Title = {A Useful Method for Model-building},
Year = 1962,
Journal = {Technometrics},
Volume = 4,
Pages = {301--318},
Keywords = {Adaptive modeling; Empirical model; Kinetic model;
Nonlinear}
I'll consider a vector response, because that is easier, but you can
generalize to matrix.
I would start with various kinds of plots of the response vs.
different poistions in the vector. With a 2 x 2 experiment and more
than 4 runs, I would make 4 such plots, one for each set of conditions.
Box and Hunter then suggest to model each line (vector or matrix)
separately. This will reduce the dimensionality of the problem to
something more managable with traditional techniques.
howe this helps. spencer graves
ZABALZA-MEZGHANI Isabelle wrote:> Hello,
>
> I wonder if anybody has some idea about how to solve my problem :
>
> I am working , I would say trough an experimental design approach (perform
> experiments, get responses, make regression, sensitivity analysis, risk
> analysis ...). The problem is now that I have to face with not only a
> response but a vector or a matrix (typically a spatial distribution of a
> physical property ... pressure). Is there any kind of approach to deal with
> that ? I don' t know to group cells together to dicrease the dimension
of
> the problem ...
>
> I have no idea and I really need some help
>
> Thanks in advance
>
> Isabelle
>
> Isabelle Zabalza-Mezghani
> Institut Fran?ais du P?trole
> France
>
> ______________________________________________
> R-help at stat.math.ethz.ch mailing list
> https://www.stat.math.ethz.ch/mailman/listinfo/r-help