Chapter Contents
Chapter Contents
Previous
Previous
Next
Next
Details of the OPTEX Procedure

Summary of Functions

Table 24.1, Table 24.2, and Table 24.3 classify the OPTEX statements and options by function.

Table 24.1: Summary of Options for Specifying the Design

Function Statement Option
Design Characteristics  
Number of design pointsGENERATEN=number
Saturated designGENERATEN=SATURATED
Augmented designGENERATEAUGMENT=SAS-data-set
Bayesian optimal designMODEL/ PRIOR=p1,p2, ...
   
Optimality Criteria  
Minimize trace of (X'X)-1GENERATECRITERION=A
Maximize |X'X|GENERATECRITERION=D
Minimize mean minimumGENERATECRITERION=U
distance to design  
Maximize mean distanceGENERATECRITERION=S
between nearest design points  
   
Model Specification  
Specify independent effectsMODELeffects
Exclude intercept termMODELeffects NOINT
Specify class variablesCLASSvariables
Static codingPROC OPTEXCODING=STATIC
Orthogonal codingPROC OPTEXCODING=ORTH
Orthogonal coding withPROC OPTEXCODING=ORTHCAN
respect to candidates only  
Suppress coding of effectsPROC OPTEXNOCODE
   
Block Specification  
Specify general covarianceBLOCKSCOVAR=SAS-data-set <options>
matrix for runs VAR=variables
Specify general covariate modelBLOCKSDESIGN=SAS-data-set <options>
Specify b blocks of size kBLOCKSSTRUCTURE=(b)k <options>
Options for block specifications  
Repeat the search n times ITER=n
Retain best m searches KEEP=m
Select initial design at random INIT=RANDOM
Select initial design in order INIT=CHAIN
   
Initial Design Characteristics  
Random and sequential methodsGENERATEINITDESIGN=PARTIAL<(m)>
Random initial designGENERATEINITDESIGN=RANDOM
Sequential initial designGENERATEINITDESIGN=SEQUENTIAL
Specify initial designGENERATEINITDESIGN=SAS-data-set
   

Table 24.2: Summary of Options for Searching for the Design

Function Statement Option
Design Search Specification  
Retain best n searchesGENERATEKEEP=n
Search n timesGENERATEITER=n
Specify candidate pointsPROC OPTEXDATA=SAS-data-set
Specify random seedPROC OPTEXSEED=number
Specify effective zeroPROC OPTEXEPSILON=\epsilon
   
Design Search Methods  
DETMAX algorithm withGENERATEMETHOD=DETMAX<(level)>
maximum excursion level  
Exchange algorithmGENERATEMETHOD=EXCHANGE
k-Exchange algorithmGENERATEMETHOD=EXCHANGE<(k)>
Sequential algorithmGENERATEMETHOD=SEQUENTIAL
Fedorov algorithmGENERATEMETHOD=FEDOROV
Modified Fedorov algorithmGENERATEMETHOD=M_FEDOROV
   

Table 24.3: Summary of Options for Examining and Saving the Design

Function Statement Option
Save the Design  
Best designOUTPUT OUT=SAS-data-set 
Specific designOUTPUT OUT=SAS-data-setNUMBER=design-number
Block variable nameOUTPUT OUT=SAS-data-setBLOCK=variable-name
Specify transfer variablesIDvariables
   
List the Design  
Design characteristicsEXAMINE 
Design pointsEXAMINEDESIGN
Information matrix X'XEXAMINEINFORMATION
Specific optimal designEXAMINENUMBER=design-number
Variance matrix (X'X)-1EXAMINEVARIANCE
Suppress all outputPROC OPTEXNOPRINT
   

Chapter Contents
Chapter Contents
Previous
Previous
Next
Next
Top
Top

Copyright © 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.