ADXADCEN: Add Center Points

Macros for the Design and Analysis of Experiments

ADXADCEN: Add Center Points

%adxadcen(ds, vlst, nc, bvar)

where

ds is the name of the SAS data set that contains the design to which you want to add centerpoints.

vlst is the list of design factors.

nc is the number of centerpoints to be added to the design.

bvar is the block variable, if any. If blocking is involved, then the design is sorted by blocks, and nc center points are added to each block.

The ADXADCEN macro adds a specified number of centerpoints to a design stored in a SAS data set. Centerpoints are runs where all factors are at their central levels (with coded values of 0). They are often added to a fractional factorial design to allow testing for lack-of-fit, and they form part of a central composite design.

Suppose you use the ADXFFD macro to create a fractional factorial design for 5 factors in 16 runs and 4 blocks and output the design to the data set named DESIGN, keeping the default factor names T1, T2, T3, T4, and T5 and the default block factor name (BLOCK). You want to add three centerpoints to each block of the design. The following statements create the fractional factorial design, output it to the data set DESIGN, and add three centerpoints to each block:

   %adxgen
   %adxff
   %adxcc
   %adxinit
   %adxffd(design,5,16,4)
   %adxadcen(design,t1 t2 t3 t4 t5,3,block)

Chapter Contents
Previous
Next
Top

ds	is the name of the SAS data set that contains the design to which you want to add centerpoints.
vlst	is the list of design factors.
nc	is the number of centerpoints to be added to the design.
bvar	is the block variable, if any. If blocking is involved, then the design is sorted by blocks, and nc center points are added to each block.