Chapter Contents |
Previous |
Next |
Macros for the Design and Analysis of Experiments |
where
ds | gives the name of the output data set for the constructed design. |
nf | gives the number of factors in the design. |
The ADXPBD macro constructs Plackett-Burman designs. These are orthogonal main-effect screening designs with a minimum number of runs, for 1 to 47 factors. The designs were first discussed by Plackett and Burman (1947).
The number of runs in these designs will be the smallest multiple of 4 that is greater than the number of factors. These are main-effects-only designs. The designs give no protection against two-factor interactions and have few (if any) degrees of freedom available for error. Thus, these designs should be used with caution. They are most appropriate in situations where a very large number of factors need to be screened and a minimum-run design is absolutely necessary.
To construct a design for 11 two-level factors and store it in the data set A, use the following statements:
%adxgen %adxff %adxinit %adxpbd(a,11)
Chapter Contents |
Previous |
Next |
Top |
Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.