Syntax


The following statements are available in PROC TRANSREG.

  • PROC TRANSREG < DATA= SAS-data-set >

  • < OUTTEST= SAS-data-set >< a-options >< o-options > ;

  • MODEL < transform(dependents < / t-options > )

    • < transform(dependents < / t-options > )... > = >

      transform(independents < / t-options > )

      < transform(independents < / t-options > )... >< / a-options > ;

  • OUTPUT < OUT= SAS-data-set >< o-options > ;

    ID variables ;

    FREQ variable ;

    WEIGHT variable ;

    BY variables ;

To use the TRANSREG procedure, you need the PROC TRANSREG and MODEL statements. To produce an OUT= output data set, the OUTPUT statement is required. PROC TRANSREG enables you to specify the same options in more than one statement. All of the MODEL statement a-options (algorithm options) and all of the OUTPUT statement o-options (output options) can also be specified in the PROC TRANSREG statement. You can abbreviate all a-options , o-options , and t-options (transformation options) to their first three letters . This is a special feature of the TRANSREG procedure and is not generally true of other SAS/STAT procedures. See Table 75.1 on page 4554.

Table 75.1: Options Available in the TRANSREG Procedure

Task

Option

Statement

Identify input data set

specifies input SAS data set

DATA=

PROC

Output data set with test statistics

specifies output test statistics data set

OUTTEST=

PROC

Input data set

specifies input observation type

TYPE=

MODEL

restarts iterations

REITERATE

MODEL

Specify method and control iterations

specifies minimum criterion change

CCONVERGE=

MODEL

specifies minimum data change

CONVERGE=

MODEL

specifies canonical dummy -variable initialization

DUMMY

MODEL

specifies maximum number of iterations

MAXITER=

MODEL

specifies iterative algorithm

METHOD=

MODEL

specifies number of canonical variables

NCAN=

MODEL

specifies singularity criterion

SINGULAR=

MODEL

Control missing data handling

METHOD=MORALS fists each model individually

INDIVIDUAL

MODEL

includes monotone special missing values

MONOTONE=

MODEL

excludes observations with missing values

NOMISS

MODEL

unties special missing values

UNTIE=

MODEL

Control intercept and CLASS variables

CLASS dummy variable name prefix

CPREFIX=

MODEL

CLASS dummy variable label prefix

LPREFIX=

MODEL

no intercept or centering

NOINT

MODEL

order of class variable levels

ORDER=

MODEL

controls output of reference levels

REFERENCE=

MODEL

CLASS dummy variable label separators

SEPARATORS=

MODEL

Control displayed output

confidence limits alpha

ALPHA=

MODEL

displays parameter estimate confidence limits

CL

MODEL

displays model specification details

DETAIL

MODEL

displays iteration histories

HISTORY

MODEL

suppresses displayed output

NOPRINT

MODEL

suppresses the iteration histories

SHORT

MODEL

displays regression results

SS2

MODEL

displays ANOVA table

TEST

MODEL

displays conjoint part-worth utilities

UTILITIES

MODEL

Control standardization

fits additive model

ADDITIVE

MODEL

do not zero constant variables

NOZEROCONSTANT

MODEL

specifies transformation standardization

TSTANDARD=

MODEL

Predicted values, residuals, scores

outputs canonical scores

CANONICAL

OUTPUT

outputs individual confidence limits

CLI

OUTPUT

outputs mean confidence limits

CLM

OUTPUT

specifies design matrix coding

DESIGN=

OUTPUT

outputs leverage

LEVERAGE

OUTPUT

does not restore missing values

NORESTOREMISSING

OUTPUT

suppresses output of scores

NOSCORES

OUTPUT

outputs predicted values

PREDICTED

OUTPUT

outputs redundancy variables

REDUNDANCY=

OUTPUT

outputs residuals

RESIDUALS

OUTPUT

Output data set replacement

replaces dependent variables

DREPLACE

OUTPUT

replaces independent variables

IREPLACE

OUTPUT

replaces all variables

REPLACE

OUTPUT

Output data set coefficients

outputs coefficients

COEFFICIENTS

OUTPUT

outputs ideal point coordinates

COORDINATES

OUTPUT

outputs marginal means

MEANS

OUTPUT

outputs redundancy analysis coefficients

MREDUNDANCY

OUTPUT

Output data set variable name prefixes

dependent variable approximations

ADPREFIX=

OUTPUT

independent variable approximations

AIPREFIX=

OUTPUT

canonical dependent variables

CDPREFIX=

OUTPUT

conservative individual lower CL

CILPREFIX=

OUTPUT

canonical independent variables

CIPREFIX=

OUTPUT

conservative-individual-upper CL

CIUPREFIX=

OUTPUT

conservative-mean-lower CL

CMLPREFIX=

OUTPUT

conservative-mean-upper CL

CMUPREFIX=

OUTPUT

METHOD=MORALS untransformed dependent

DEPENDENT=

OUTPUT

liberal -individual-lower CL

LILPREFIX=

OUTPUT

liberal-individual-upper CL

LIUPREFIX=

OUTPUT

liberal-mean-lower CL

LMLPREFIX=

OUTPUT

liberal-mean-upper CL

LMUPREFIX=

OUTPUT

residuals

RDPREFIX=

OUTPUT

predicted values

PPREFIX=

OUTPUT

redundancy variables

RPREFIX=

OUTPUT

transformed dependents

TDPREFIX=

OUTPUT

transformed independents

TIPREFIX=

OUTPUT

Output data set macros

creates macro variables

MACRO

OUTPUT

Output data set details

dependent and independent approximations

APPROXIMATIONS

OUTPUT

canonical correlation coefficients

CCC

OUTPUT

canonical elliptical point coordinate

CEC

OUTPUT

canonical point coordinates

CPC

OUTPUT

canonical quadratic point coordinates

CQC

OUTPUT

approximations to transformed dependents

DAPPROXIMATIONS

OUTPUT

approximations to transformed independents

IAPPROXIMATIONS

OUTPUT

elliptical point coordinates

MEC

OUTPUT

point coordinates

MPC

OUTPUT

quadratic point coordinates

MQC

OUTPUT

multiple regression coefficients

MRC

OUTPUT

The rest of this section provides detailed syntax information for each of the preceding statements, beginning with the PROC TRANSREG statement. The remaining statements are described in alphabetical order.




SAS.STAT 9.1 Users Guide (Vol. 7)
SAS/STAT 9.1 Users Guide, Volumes 1-7
ISBN: 1590472438
EAN: 2147483647
Year: 2004
Pages: 132

flylib.com © 2008-2017.
If you may any questions please contact us: flylib@qtcs.net