BookmarkSubscribeRSS Feed
acordes
Rhodochrosite | Level 12

I have 700k distinct groupby counts. 

 

If I use the input={} statement it generates correctly the columns _frequency_ and _position_, but the _cumfreq_ is not what I want top have. I would like to accumulate the freq within a contract (numero_operacion).  My first groupbyinfo run relates to this attempt which at least runs quickly and gives correct freq and position counts. 

 

When I try to use the groupby={} option in the table{} statement, then it takes en eternity to finish. 

It's not so big to justify this amount of time, but probably I'm missing out something. 

 

I don't know how to use and combine, vars{}, inputs{} and groupby{} wisely. And I am lost to fine-tune by setting algorithm2, groupbylimit groupbyorder,...

 

Can someone help out and provide a complex example. the examples from the sas site should cover more sophisticated examples. 

 

By the way, I know how to resolve my problem with alternative approaches.

But I want to truely understand the groupbyinfo action. 

thanks a lot.

 

proc cas;
   session mysession;
                             simple.groupByInfo /                          /* 1 */
               includeDuplicates=true,
               minFrequency=1,
                              generatedcolumns={'frequency', 'position', 'cumfreq'},
                              groupByLimit=100M,
                              nworkerthreads=8,
               noVars=true,
                              algorithm2=true, 
                              journaltrace=true,
                              inputs={'Numero_operacion', 'JourneyName'},
               casOut={name="testa_dup", replace=true, CASLIB="mkt"},

               table={
/*                                                       vars={'Numero_operacion', 'JourneyName'}, groupBy={'_date'},  */
                                                           groupByMode="redistribute", orderBy='fecha_envio',
/*                          orderBy="fecha_envio", */
                              name="FUNNEL_REN", CASLIB="mkt"
                                            ,computedVars={
                                            name="_date"}, 
                                            computedVarsProgram="_date=put(datepart(fecha_envio), monyy.);"
};
run;

proc cas;
   table.fetch /                                  
      format=True, maxrows=100, 
/*       fetchVars={ */
/*          "_score_", 'Numero_operacion', 'JourneyName'}, */
      table={name="testa_dup", caslib="mkt", where="Numero_operacion in ('01RN40000930' '01RN27000014')"};
run;


proc cas;
   session mysession;
                             simple.groupByInfo /                          /* 1 */
               includeDuplicates=true,
               minFrequency=1,
                              generatedcolumns={'frequency', 'position', 'cumfreq'},
                              groupByLimit=100M,
                              details=true,
                              nworkerthreads=8,
               noVars=true,
                              algorithm2=true, 
                              journaltrace=true,
                              inputs={'Numero_operacion', 'JourneyName'},
               casOut={name="testa_dup", replace=true, CASLIB="mkt"},

               table={groupby={'Numero_operacion', 'JourneyName'}, groupByMode="redistribute", 
/*                          orderBy='fecha_envio', */
/*                          orderBy="fecha_envio", */
                              name="FUNNEL_RE1", CASLIB="mkt"
                                            ,computedVars={
                                            name="_date"}, 
                                            computedVarsProgram="_date=put(datepart(fecha_envio), monyy.);"
};
run;

hackathon24-white-horiz.png

The 2025 SAS Hackathon has begun!

It's finally time to hack! Remember to visit the SAS Hacker's Hub regularly for news and updates.

Latest Updates

How to Concatenate Values

Learn how use the CAT functions in SAS to join values from multiple variables into a single value.

Find more tutorials on the SAS Users YouTube channel.

SAS Training: Just a Click Away

 Ready to level-up your skills? Choose your own adventure.

Browse our catalog!

Discussion stats
  • 0 replies
  • 611 views
  • 0 likes
  • 1 in conversation