Chapter Contents |
Previous |
Next |
The UNIVARIATE Procedure |
Tip: | You can save percentiles that are not automatically computed. |
Tip: | You can use multiple OUTPUT statements to create several OUT= data sets. |
Main discussion: | Output Data Set |
Featured in: | Examining the Data Distribution and Saving Percentiles , Creating an Output Data Set with Multiple Analysis Variables , and Creating Schematic Plots and an Output Data Set with BY Groups |
OUTPUT
<OUT=SAS-data-set> statistic-keyword-1=name(s)
<...statistic-keyword-n=name(s)> <percentiles-specification> ; |
Options |
Default: | DATAn |
Descriptive statistic keywords | |||
CSS | CV | KURTOSIS | |
MAX | MEAN | N | |
MIN | MODE | RANGE | |
NMISS | NOBS | STDMEAN | |
SKEWNESS | STD | USS | |
SUM | SUMWGT | VAR | |
Quantile statistic keywords | |||
MEDIAN | P1 | P5 | |
P10 | P90 | P95 | |
P99 | Q1 | Q3 | |
QRANGE | |||
Robust statistic keywords | |||
GINI | MAD | QN | |
SN | STD_GINI | STD_MAD | |
STD_QN | STD_QRANGE | STD_SN | |
Hypothesis testing keywords | |||
NORMAL | PROBN | MSIGN | |
PROBM | SIGNRANK | PROBS | |
T | PROBT |
See SAS Elementary Statistics Procedures and Statistical Computations for the keyword definitions and statistical formulas.
To store the same statistic for several analysis variables, specify a list of names. The order of the names corresponds to the order of the analysis variables in the VAR statement. PROC UNIVARIATE uses the first name to create a variable that contains the statistic for the first analysis variable, the next name to create a variable that contains the statistic for the second analysis variable, and so on. If you do not want to output statistics for all the analysis variables, specify fewer names than the number of analysis variables.
PCTLPTS=percentile(s) PCTLPRE=prefix-name(s) <PCTLNAME=suffix-name(s)> |
Range: | any decimal numbers between 0 and 100, inclusive |
Example: | To compute the 50th, 95th, 97.5th, and
100th percentiles, submit the statement
output pctlpre=P_ pctlpts=50,95 to 100 by 2.5; |
Interaction: | PROC UNIVARIATE creates a variable name by combining the PCTLPRE= value and either suffix-name or (if you omit PCTLNAME= or if you specify too few suffix-name(s)) the PCTLPTS= value. |
Requirement: | You must specify PCTLPRE= to supply prefix names for the variables that contain the PCTLPTS= percentiles. |
Interaction: | If the number of PCTLNAME= values is fewer than the number of percentile(s) or if you omit PCTLNAME=, PROC UNIVARIATE uses percentile as the suffix to create the name of the variable that contains the percentile. For an integer percentile, PROC UNIVARIATE uses percentile. For a noninteger percentile, PROC UNIVARIATE truncates decimal values of percentile to two decimal places and replaces the decimal point with an underscore. |
Interaction: | If either the prefix and suffix name combination or the prefix and percentile name combination is longer than 32 characters, PROC UNIVARIATE truncates the prefix name so that the variable name is 32 characters. |
Saving Percentiles Not Automatically Computed |
proc univariate data=score; var Test1 Test2; output out=pctls pctlpts=20 40 pctlpre=Test1_ Test2_ pctlname=P20 P40; run;PROC UNIVARIATE saves the 20th and 40th percentiles for Test1 and Test2 in the variables Test1_P20, Test2_P20, Test1_P40, and Test2_P40.
Using the BY Statement with the OUTPUT Statement |
Chapter Contents |
Previous |
Next |
Top of Page |
Copyright 1999 by SAS Institute Inc., Cary, NC, USA. All rights reserved.