The OPTEX Procedure |
You use the OUTPUT statement to save a design in an output data set. By default, the saved design is the best design found. You specify the data set name as follows:
The following options can be used:
specifies the name to be given to the blocking variable in the output data set. The default name is BLOCK. You can use this option in conjunction with a STRUCTURE= option in the BLOCKS statement. See Example 10.7 for an example.
selects a design to output by specifying its design-number. Designs are ordered by the value of the efficiency criterion that is being optimized. Thus, a design-number of 1 corresponds to the best design found, a design-number of 2 corresponds to the second best design, and so on. The default design-number is 1. To modify the number of designs created, see the ITER= option.
Alternatively, you can specify one of the following:
selects the design that has the highest D-efficiency value.
selects the design that has the highest A-efficiency value.
selects the design that has the highest G-efficiency value.
selects the design that has the minimum average standard error for prediction.
These options can be used to find designs that are efficient for more than one criterion For example, you can use the default CRITERION=D option in the GENERATE statement with the NUMBER=GBEST option in the OUTPUT statement to find the D-optimal design that has maximal G-efficiency. In fact, this is the best way to use the OPTEX procedure to find G-efficient designs; see the section G- and I-optimality for more details.
Copyright © SAS Institute, Inc. All Rights Reserved.