ADXSCD: Construct Simplex-Centroid Designs

Macros for the Design and Analysis of Experiments

ADXSCD: Construct Simplex-Centroid Designs

%adxscd(ds, vlst, m)

where

ds is the name of the SAS data set for the constructed design. This data set contains variables for design factors. Default names are used. The default names for design factors are T1, T2, T3, and so on.

vlst is a list of mixture components.

m is the degree of the model.

The ADXSCD macro can be used to construct simplex-centroid designs of any degree m. These designs are efficient for fitting an m^th-degree polynomial model to mixture data. A simplex-centroid design of degree m is composed of mixtures with only one factor present, mixtures with two factors present in equal amounts, mixtures with three equal factors, and so on up to m; and the mixture that contains all factors in equal amounts.

The design data set includes the levels of the mixture components for each point in the design. In addition, the data set includes a variable, DIMEN, which gives the dimension of the face from which the corresponding centroid is computed. A 0-dimensional centroid is a vertex, a 1-dimensional centroid is computed from an edge, and so on.

For example, suppose you want a design for an experiment on blends of three fruit juices: watermelon, pineapple, and orange. To construct a simplex-centroid design of order 2, submit the following statements:

   %adxgen
   %adxmix
   %adxinit
   %adxscd(juice,watermel pineappl orange,2)

The design is output to the JUICE data set, which contains the variables WATERMEL, PINEAPPL, and ORANGE. The values of these variables give the proportions of each juice for runs in the experimental design.

Chapter Contents
Previous
Next
Top

ds	is the name of the SAS data set for the constructed design. This data set contains variables for design factors. Default names are used. The default names for design factors are T1, T2, T3, and so on.
vlst	is a list of mixture components.
m	is the degree of the model.