Special SAS Data Sets

Introduction to Special SAS Data Sets

Many SAS/STAT procedures create SAS data sets containing various statistics. Some of these data sets are organized according to certain conventions that allow them to be read by a SAS/STAT procedure for further analysis. Such specially organized data sets are recognized by the TYPE= attribute of the data set.

For example, the CORR procedure (refer to the SAS Procedures Guide) can create a data set with the attribute TYPE=CORR containing a correlation matrix. This TYPE=CORR data set can be read by the REG or FACTOR procedure, among others. If the original data set is large, using a special SAS data set in this way can save a great deal of computer time by avoiding the recomputation of the correlation matrix in each of several analyses.

As another example, the REG procedure can create a TYPE=EST data set containing estimated regression coefficients. If you need to make predictions for new observations, you can have the SCORE procedure read both the TYPE=EST data set and a data set containing the new observations. PROC SCORE can then compute predicted values or residuals without repeating the entire regression analysis. See Chapter 57, "The SCORE Procedure," for an example.

Many procedures provide options to produce output data sets of TYPE=null; you can also create TYPE=null data sets from any table generated by a procedure by using the Output Delivery System (ODS); see Chapter 15, "Using the Output Delivery System," for more information.

A special SAS data set may contain different kinds of statistics. A special variable called _TYPE_ is used to distinguish the various statistics. For example, in a TYPE=CORR data set, an observation in which _TYPE_='MEAN' contains the means of the variables in the analysis, and an observation in which _TYPE_='STD' contains the standard deviations. Correlations appear in observations with _TYPE_='CORR'. Another special variable, _NAME_, is needed to identify the row of the correlation matrix. Thus, the correlation between variables X and Y would be given by the value of the variable X in the observation for which _TYPE_='CORR' and _NAME_='Y', or by the value of the variable Y in the observation for which _TYPE_='CORR' and _NAME_='X'.

You can create special SAS data steps directly in a DATA step by specifying the TYPE= option in parentheses after the data set name in the DATA statement. See Example A.2 for an example.

The special data sets created by SAS/STAT procedures can generally be used directly by other procedures without modification. However, if you create an output data set with PROC CORR and use the NOCORR option to omit the correlation matrix from the OUT= data set, you need to set the TYPE= option either in parentheses following the OUT= data set name in the PROC CORR statement or in parentheses following the DATA= option in any other procedure that recognizes the special TYPE= attribute. In either case, the TYPE= option should be set to COV, CSSCP, or SSCP according to what type of matrix is stored in the data set and what data set types are accepted as input by the other procedures you plan to use. If you do not follow these steps and you use the TYPE=CORR data set with no correlation matrix as input to another procedure, the procedure may issue an error message indicating that the correlation matrix is missing from the data set.

If you use a DATA step with a SET statement to modify a special SAS data set, you must specify the TYPE= option in the DATA statement. The TYPE= attribute of the data set in the SET statement is not automatically copied to the data set being created.

You can determine the TYPE= attribute of a data set by using the CONTENTS procedure (see Example A.1 and refer to the SAS Procedures Guide for details).

Table A.1 summarizes the TYPE= data sets that can be used as input to SAS/STAT procedures and the TYPE= data sets that are created by SAS/STAT procedures. The essential parts of the statements each procedure uses to create its output data set or data sets are shown.

Formulas useful for illustrating differences between corrected and uncorrected matrices in some special SAS data sets are shown in the "Definitional Formulas" section.

Table A.1: SAS/STAT Procedures and Types of Data Sets

Procedure	Input Data Sets TYPE= as shown^*	Output Data Sets (TYPE=null or as shown)	Created by Statement and Specification
ACECLUS	INITIAL= INPUT= data set may be of type: ACE, CORR, COV, SSCP, UCORR, UCOV	ACE	PROC ACECLUS OUTSTAT= PROC ACECLUS OUT=
ANOVA			PROC ANOVA OUTSTAT=
CALIS	CORR, COV, FACTOR, RAM, SSCP, UCORR, UCOV, WEIGHT	CORR COV EST UCORR UCOV RAM WEIGHT	PROC CALIS OUTSTAT= PROC CALIS COV OUTSTAT= PROC CALIS OUTEST= PROC CALIS NOINT OUTSTAT= PROC CALIS NOINT COV OUTSTAT= PROC CALIS OUTRAM= PROC CALIS OUTWGT=
CANCORR	CORR, COV, FACTOR, SSCP, UCORR, UCOV	CORR UCORR	PROC CANCORR OUTSTAT= PROC CANCORR NOINT OUTSTAT= PROC CANCORR OUT=
CANDISC	CORR, COV, SSCP, CSSCP	CORR	PROC CANDISC OUTSTAT= PROC CANDISC OUT=
CATMOD	EST	EST	RESPONSE / OUTEST= RESPONSE /OUT=
CLUSTER	DISTANCE	TREE	PROC CLUSTER OUTTREE=
CORRESP			PROC CORRESP OUTC= PROC CORRESP OUTF=
DISCRIM	CORR, COV, SSCP, CSSCP, LINEAR, QUAD, MIXED	LINEAR QUAD MIXED CORR	PROC DISCRIM POOL=YES OUTSTAT= PROC DISCRIM POOL=NO OUTSTAT= PROC DISCRIM POOL=TEST OUTSTAT= PROC DISCRIM METHOD=NPAR OUTSTAT= PROC DISCRIM OUT= PROC DISCRIM OUTCROSS= PROC DISCRIM OUTD= PROC DISCRIM TESTOUT= PROC DISCRIM TESTOUTD=
FACTOR	ACE, CORR, COV, FACTOR, SSCP, UCORR, UCOV	FACTOR	PROC FACTOR OUTSTAT= PROC FACTOR OUT=
FASTCLUS			PROC FASTCLUS OUT= PROC FASTCLUS OUTSEED= PROC FASTCLUS OUTSTAT= PROC FASTCLUS MEAN=
FREQ			TABLES OUT= OUTPUT OUT=
GENMOD			OUTPUT OUT=
GLM			PROC GLM OUTSTAT= LSMEANS / OUT= OUTPUT OUT=
GLMMOD			PROC GLMMOD OUTDESIGN= PROC GLMMOD OUTPARM=
INBREED			PROC INBREED OUTCOV=
KRIGE2D			PROC KRIGE2D OUTEST= PROC KRIGE2D OUTNBHD=
LATTICE
LIFEREG		EST	PROC LIFEREG OUTEST= OUTPUT OUT=
LIFETEST			PROC LIFETEST OUTSURV= PROC LIFETEST OUTTEST=
LOGISTIC		EST	PROC LOGISTIC OUTEST= OUTPUT OUT= MODEL / OUTROC=
MDS			PROC MDS OUT= PROC MDS OUTFIT= PROC MDS OUTRES=
MIXED			MODEL OUTPRED= MODEL OUTPREDM= PRIOR OUT= PRIOR OUTG= PRIOR OUTGT=
MODECLUS	DISTANCE		PROC MODECLUS OUT= PROC MODECLUS OUTCLUS= PROC MODECLUS OUTSUM=
MULTTEST			PROC MULTTEST OUT= PROC MULTTEST OUTPERM= PROC MULTTEST OUTSAMP=
NESTED
NLIN		EST	PROC NLIN OUTEST= OUTPUT OUT=
NPAR1WAY			OUTPUT OUT=
ORTHOREG		EST	PROC ORTHOREG OUTEST=
PHREG		EST	PROC PHREG OUTEST= BASELINE OUT= OUTPUT OUT=
PLAN			OUTPUT OUT=
PLS			OUTPUT OUT=
PRINCOMP	ACE, CORR, COV, EST, FACTOR, SSCP, UCORR, UCOV	CORR COV UCORR UCOV	PROC PRINCOMP OUTSTAT= PROC PRINCOMP COV OUTSTAT= PROC PRINCOMP NOINT OUTSTAT= PROC PRINCOMP NOINT COV OUTSTAT= PROC PRINCOMP OUT=
PRINQUAL			PROC PRINQUAL OUT=
PROBIT		EST	PROC PROBIT OUTEST= OUTPUT OUT=
REG	CORR, COV, SSCP, UCORR, UCOV	EST SSCP	PROC REG OUTEST= PROC REG OUTSSCP= OUTPUT OUT=
RSREG			PROC RSREG OUT= RIDGE OUTR=
SCORE	SCORE= data set can be of any type		PROC SCORE OUT=
SIM2D			PROC SIM2D OUTSIM=
SURVEYSELECT			PROC SURVEYSELECT OUT= PROC SURVEYSELECT OUTSORT=
STDIZE			PROC STDIZE OUT= PROC STDIZE OUTSTAT=
STEPDISC	CORR, COV, SSCP, CSSCP
TRANSREG			PROC TRANSREG OUTTEST= OUTPUT OUT=
TREE	TREE		PROC TREE OUT=
TTEST
VARCLUS	CORR, COV, FACTOR, SSCP, UCORR, UCOV	CORR UCORR TREE	PROC VARCLUS OUTSTAT= PROC VARCLUS NOINT OUTSTAT= PROC VARCLUS OUTTREE=
VARCOMP
VARIOGRAM			PROC VARIOGRAM OUTDISTANCE= PROC VARIOGRAM OUTPAIR= PROC VARIOGRAM OUTVAR=

^*If no TYPE= is shown, the procedure does not recognize any special data set types except possibly to issue an error message for inappropriate values of TYPE=.

Chapter Contents
Previous
Next
Top