Calculating Percentiles with SAS

In SAS, you can calculate percentiles using PROC UNIVARIATE procedure.

Important options used for calculating percentile in PROC UNIVARIATE
  1. PCTLPTS : Specifies percentile levels
  2. PCTLPRE : Specifies one or more prefixes to create the variable names for the variables that contain the PCTLPTS= percentiles.
  3. PCTLNAME : Specifies one or more suffixes to create the variable names for the variables that contain the PCTLPTS= percentiles.
Example 1 : 
Percentile calculation for a variable
proc univariate data = abcd noprint;
var a;
output out=outdata PCTLPTS = 99.5 PCTLPRE = a;
Example 2 : 

Percentile calculation for multiple variables
proc univariate data = abcd noprint;
var a b c;
output out=outdata PCTLPTS = 99 PCTLPRE = a b c PCTLNAME = _P99;

2 Important Points : 

  1. If you specify 3 variables in var statement (var a b c) and only 1 prefix in PCTPRE, SAS will create percentile for only 1 variable that is mentioned first in the var statement.
  2. If the number of PCTLNAME= values is fewer than the number of percentiles or if you omit PCTLNAME=, PROC UNIVARIATE uses the percentile as the suffix to create the name of the variable that contains the percentile.

Example 3 : 

Series of percentile levels
proc univariate data = abcd noprint;
var c;
output out=outdata PCTLPTS = 0 to 5 by 0.5 , 95 to 100 by 0.5 PCTLPRE = P;
Related Posts
About Author:

Deepanshu founded ListenData with a simple objective - Make analytics easy to understand and follow. He has over 8 years of experience in data science. During his tenure, he has worked with global clients in various domains like Banking, Insurance, Telecom and Human Resource.

1 Response to "Calculating Percentiles with SAS"
  1. Hi ,I am an avid follower of your posts and it has helped a lot in clearing and understanding certain concepts for first of all a BIG BIG THANKS for posting these articles.
    For percentiles I would like to share one article which I came across which shares the secret of getting percentile of multiple variable at one go, and also some custom percentiles. Have a look and let me know your thoughts on the same..


Next → ← Prev
Love this Post? Spread the Word!