Chapter Contents |
Previous |
Next |
Macros for the Design and Analysis of Experiments |
where nf gives the number of factors you want to study, where . This macro lists available designs and is useful when you do not know the exact parameters of the central composite design you want to construct.
The ADXPCC macro lists designs of various sizes that may or may not involve blocking. In each design, the values of the axial extreme are chosen so the design is rotatable, and the number of centerpoints is chosen so the design is as nearly orthogonal as possible and orthogonally blocked if blocking is involved. For details on these design characteristics, refer to Myers (1976).
In addition to listing designs, the output from the ADXPCC macro gives statements to construct each design using the ADXCCD macro. In the statements shown on the output, you need to fill in a SAS data set name for the value *data-set-name*. When you have chosen a design, you can simply submit the corresponding statement to construct the design.
When choosing a design, consider the overall number of runs required, the number of runs required in each block, and the number of centerpoints required.
The statements below produce a list of four designs, which differ depending on whether they are based on half-replicates, and whether they involve blocking. The output also includes the ADXCCD macro statements for each of these designs.
%adxgen %adxff %adxcc %adxpcc(5)
Chapter Contents |
Previous |
Next |
Top |
Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.