🚀 go-pugleaf

Crossposted to sci.engr.chem  and sci.engr.control


Hello Engineers,

I tried to ask the mathematicians in sci.math.stat about experimental 
planning, but I did get answers on a *very* theoretical level.

I would like to use a test plan optimized so that the covariance 
between the variables is as low as possible, becouse I have no 
information about the nonlinear modelstructure  that will describe
the system in the best way.  What do You think about that choise?

If the modelstructure is known a good criteria is the D-critera
( Some feedback from Don Ramirez Univ of Virginia:
>The D-optimality criterion is based on the product of the
>eigenvalues of (X'X)**(-1) and is small when only one of these
>eigenvalues are small.
>	A new criterion is based on the mean standard error of a random
>test of u'b; that is the expectation over the unit sphere of a random
>linear inference. This problem has been recently solved by Charles Dunkl
> and Don Ramirez of the University of Virginia and will appear in The
>Transactions of Mathematical Software ACM. 
)


MY QUESTIONS:

1. References or comments on giving low covariance between independent 
   variables in a testplan high priority ?

2. What's Your criteria when planning tests for unknown modelstructures? 



Post or mail, whatever You prefer.
I will summarize Your comments if they are many and 
not posted to the newsgroup.

All comments appreciated,
Staffan

        
 ----------------------------------------------------------------------------
 -               Project researcher / Environmental Engineering             -   
 ----------------------------------------------------------------------------
 -  Staffan Asplund                  -              email sasplund@abo.fi   -
 -  Process Design Lab.              -              phone +358-21-654552    -
 -  Abo Akademi University           -              ph. home  921-882744    -
 -  Finland                          -              fax   +358-21-654791    -
 ----------------------------------------------------------------------------