proc HPBIN

[asifgeneral]

Hello,

Can proc HPBIN handle categorical variables? If not what is the most efficient ways of collapsing categorical variables.

Thanks,   

[PeterClemmensen]

Yes, that should not be a problem. What does your data look like?

[asifgeneral]

Thanks for your response 

Here is the code that I am trying.

proc hpbin data=pg1.hmeq output=work.woebinn numbin=3 winsor;
input mortdue;
input yoj;
input job;
ods output mapping=mapping
run;
 

I am getting the output as expected for the two interval variables, but for job... 

ERROR: JOB is class variable, only interval variable is supported.

[Ksharp]

https://communities.sas.com/t5/Mathematical-Optimization/Trying-to-use-PROC-OPTMODEL-for-monotonic-s...

There is a my paper 

“Get Better Weight of Evidence for Scorecards Using a Genetic Algorithm.”

https://www.sas.com/content/dam/SAS/support/en/sas-global-forum-proceedings/2018/1808-2018.pdf

To bin category variable to make IV bigger.

[asifgeneral]

Thanks for your response!

Being a novice in this field of credit scoring, I will take some time to digest the information. 🙂 

Will get back in case of any doubts. Thanks for your time and helping me out with the resources.

Regards,   

[chapidi99]

Hello,

Why would you like to bin categorical variables? The categorical variables will already have categories. It is meaningful to categorise continious variables to create groups to measure the strength of the groups.

For example, you have the column House_Status with the categories "Own_House", 'Rented' and "Missing". How would you further categorise them. Even if you assign a numeric value as, for example,  Own_House = 0, Rented = 1 and Missing = 2 further categorising them wouldn't make any sense.

Many thanks,

Kiran Chapidi