A Central Composite Design of Experiments
H = scidoe_ccdesign(nbvar) H = scidoe_ccdesign(nbvar,"Name",value,...)
a 1-by-1 matrix of doubles, positive integer, the number of variables of the experiment
a 1-by-1 matrix of strings. The options are "type", "alpha" and "center".
the value
This function produces a Central Composite Design (CCD) of experiments. A CCD is composed by :
a factorial block design,
a block of axial (star) points and
a matrix of center points (zeros).
The available options are the following.
"type" : "circumscribed" (default), "inscribed" or "faced".
"alpha" : "orthogonal" (default) or "rotatable".
"center" : a 1-by-2 row vector of doubles, the number of center points in each block of the design.
The "circumscribed" and "inscribed" can be rotatable designs, but "faced" cannot. For "faced" CCD, alpha =1. If "type" is specified, while "alpha" is not, then default value is "orthogonal".
If the design is "orthogonal",
alpha = sqrt(nbvar*(1+(nao/na))/(1+(nco/nc))),
nc : the number of factorial points,
nco : the number of center points added to the factorial design,
na : the number of axial points,
nao : the number of center points added to the axial points.
If the design is "rotatable",
alpha = nc^1/4,
If the design is "faced", then alpha=1.
The user can specify the number of center points in each block of the design (factorial and axial). If "Parameter" is [2 3], then there are 2 center points in the factorial block and 3 points in the axial points block. Default value is [4 4] (a total of 8 center points in the CCD).
There is no strict order in the input of the above options. For example,
scidoe_ccdesign(2,"type","inscribed","center",[2 3])
scidoe_ccdesign(2,"center",[2 3],"type","inscribed")
George E. P. Box, J. Stuart Hunter, William G. Hunter, "Statistics for experimenters Design,Innovation and Discovery", Second Edition, 2005
http://en.wikipedia.org/wiki/Central_composite_design
http://rgm2.lab.nig.ac.jp/RGM2/func.php?rd_id=rsm:ccd
http://www.mathworks.com/help/toolbox/stats/f56635.html
http://www.itl.nist.gov/div898/handbook/pri/section3/pri3361.htm