BookmarkSubscribeRSS Feed
☑ This topic is solved. Need further help from the community? Please sign in and ask a new question.
UPRETIGOPI
Obsidian | Level 7

Hi dear all

I am new in this community. I am seeking some help on BASE SAS code development for credit risk modeling  using logistic regression. If you any one of you can share your base sas codes for modeling PD / LGD/EAD  on any fictitious credit data and the idea about Stress Testing, I would greatly appreciate that. You may have written papers also on development of credit scoring model using base SAS, that would be a great help too. Thanks

 

Chandra

 

1 ACCEPTED SOLUTION

Accepted Solutions
Reeza
Super User

If you search LexJansen.com for credit risk modeling you'll find user written papers but no full case studies as data is typically proprietary. 

 

 

View solution in original post

4 REPLIES 4
SASKiwi
PROC Star

Unfortunately code developed to do credit risk modelling is typically proprietary and confidential to the companies using it so it is unlikely anyone can share it with you. Also this code tends to be very customised so is unlikely to be very useful to you anyway. Every country has their own rules regarding the regulatory capital that is calculated using EADs, PDs and LGDs so you can't reuse calculations developed for another country's rules.

 

Also In my experience, to develop credit risk models you need experts with the appropriate statistical modelling knowledge and expertise. If you don't have enough modelling expertise in your team I would raise that issue with your management.

PaigeMiller
Diamond | Level 26

Plenty of examples of using Logistic regression

https://documentation.sas.com/doc/en/pgmsascdc/9.4_3.4/statug/statug_logistic_examples.htm

 

None of them are credit risk modeling, but if you can understand these examples, the understanding also applies to credit risk modeling.

--
Paige Miller
Reeza
Super User

If you search LexJansen.com for credit risk modeling you'll find user written papers but no full case studies as data is typically proprietary. 

 

 

UPRETIGOPI
Obsidian | Level 7

Thanks Reeza 

I got few useful docs

Discussion stats
  • 4 replies
  • 1481 views
  • 3 likes
  • 4 in conversation