BookmarkSubscribeRSS Feed
Ameet
Calcite | Level 5

Hi guys,


First of all let me clarify that I am executing my code in SAS version 8.2


Have a look at the 2 scenarios below. I am basically trying to eliminate exact duplicate records from dataset "abc", keeping/dropping certain variables and also applying a filter. I understand why the first first proc sort on data abc does not delete duplicates. It is definitely because the "by" variables used do not ensure that exact duplicate records get arranged in a sequential order and hence do not get deleted.


What I do not understand is the output of the second proc sort on data "def". In both the scenarios this second proc sort is exactly the same. Why then does it not eliminate duplicates in scenario 1 whereas delete in scenario 2?


I am pretty sure I am missing something really basic here Smiley Happy


Scenario 1:

proc sort data = abc (keep = var1 var2 var3 var4 var5 var6 var7 var8)

            out = def (drop = var5) noduprec;

   by var1 var2 var3 var4;

   where upcase(compbl(var5)) = "SOME TEXT" and var6 = 9999;

run;

NOTE: 0 duplicate observations were deleted.

NOTE: There were 43 observations read from the data set ABC.

      WHERE (UPCASE(COMPBL(var5))='SOME TEXT') and (var6=9999);

NOTE: The data set WORK.DEF has 43 observations and 7 variables.

NOTE: PROCEDURE SORT used:

      real time           0.08 seconds

      cpu time            0.08 seconds

proc sort data = def  out= ghi noduprec;

   by var1 var2 var3;

run;

NOTE: Input data set is already sorted; it has been copied to the output data set.

NOTE: There were 43 observations read from the data set WORK.DEF

NOTE: The data set WORK.GHI has 43 observations and 7 variables.

NOTE: PROCEDURE SORT used:

      real time           0.00 seconds

      cpu time            0.00 seconds

Scenario 2:

proc sort data = abc (keep = var1 var2 var3 var4 var5 var6 var7 var8)

            out = def (drop = var5) ;  noduprec not used here

   by var1 var2 var3 var4;

   where upcase(compbl(var5)) = "SOME TEXT" and var6 = 9999;

run;

NOTE: There were 43 observations read from the data set ABC.

      WHERE (UPCASE(COMPBL(var5))='SOME TEXT') and (var6=9999);

NOTE: The data set WORK.DEF has 43 observations and 7 variables.

NOTE: PROCEDURE SORT used:

      real time           0.08 seconds

      cpu time            0.08 seconds

proc sort data = def  out= ghi noduprec;

   by var1 var2 var3;

run;

NOTE: 5 duplicate observations were deleted.

NOTE: There were 43 observations read from the data set WORK.DEF

NOTE: The data set WORK.GHI has 38 observations and 7 variables.

NOTE: PROCEDURE SORT used:

      real time           0.01 seconds

      cpu time            0.01 seconds

4 REPLIES 4
Patrick
Opal | Level 21

I believe you're not showing us the full picture here. How else would it be possible that you're dropping "VAR1" for output data set "def" but then can sort by "VAR1" on input data set "DEF"?

proc sort data = abc (keep = var1 var2 var3 var4 var5 var6 var7 var8)

            out = def (drop = var1) noduprec;

   by var1 var2 var3 var4;

   where upcase(compbl(var5)) = "SOME TEXT" and var6 = 9999;

run;

NOTE: 0 duplicate observations were deleted.

NOTE: There were 43 observations read from the data set ABC.

      WHERE (UPCASE(COMPBL(var5))='SOME TEXT') and (var6=9999);

NOTE: The data set WORK.DEF has 43 observations and 10 variables.

NOTE: PROCEDURE SORT used:

      real time           0.08 seconds

      cpu time            0.08 seconds

proc sort data = def  out= ghi noduprec;

   by var1 var2 var3;

run;

Ameet
Calcite | Level 5

Hi Patrick,

You are absolutely correct. Problems with data anonymization!! I replaced the dataset names and variables names from the actual dataset/variables names that I was using in production data. I have now updated my question text....I also realized that my log note had 10 variables whereas the input itself had 8 here Smiley Happy

sorry for the trouble Smiley Sad

art297
Opal | Level 21

Without seeing the actual data one can only guess! However, your main problem is likely because you are using NODUPREC. In all of my years of using SAS I have never seen a situation where that would be advisable.

Compare the definitions of NODUPREC and NODUPKEY in the documentation (e.g., Base SAS(R) 9.2 Procedures Guide),

I ALWAYS use NODUPKEY.

Doc_Duke
Rhodochrosite | Level 12

Ameet,

This tip is the reason Arthur (and most of us) don't use NODUPREC:

"Because NODUPRECS checks only consecutive observations, some nonconsecutive duplicate observations might remain in the output data set. You can remove all duplicates with this option by sorting on all variables."

If you have sorted by all variables, you might as well use NODUPKEY.

sas-innovate-2024.png

Don't miss out on SAS Innovate - Register now for the FREE Livestream!

Can't make it to Vegas? No problem! Watch our general sessions LIVE or on-demand starting April 17th. Hear from SAS execs, best-selling author Adam Grant, Hot Ones host Sean Evans, top tech journalist Kara Swisher, AI expert Cassie Kozyrkov, and the mind-blowing dance crew iLuminate! Plus, get access to over 20 breakout sessions.

 

Register now!

What is Bayesian Analysis?

Learn the difference between classical and Bayesian statistical approaches and see a few PROC examples to perform Bayesian analysis in this video.

Find more tutorials on the SAS Users YouTube channel.

Click image to register for webinarClick image to register for webinar

Classroom Training Available!

Select SAS Training centers are offering in-person courses. View upcoming courses for:

View all other training opportunities.

Discussion stats
  • 4 replies
  • 921 views
  • 4 likes
  • 4 in conversation