Programming the statistical procedures from SAS

forecasting delinquency

Accepted Solution Solved
Reply
Valued Guide
Posts: 858
Accepted Solution

forecasting delinquency

Good Afternoon All,

 

I have current data, along with the history for credit card balances.  I have grouped these by month and delinquency buckets(1 cycle = 30 days past due, 2 cycle = 60 days past due etc.)

 

There is an old process that I've inherited to forecast delinquencies, it is an markov chain that I don't like how it's written so I'm going to try to do it over using IML.  I'm also going to do a moving average. 

 

Why I'm reaching out to the group is to find out if someone has experience with an exercise like this, if they have a certain model or procedure that they thought fit this kind of data best.

 

Thank You,

 

Mark


Accepted Solutions
Solution
‎04-03-2017 10:05 AM
SAS Super FREQ
Posts: 3,555

Re: forecasting delinquency

[ Edited ]

I cannot speak to "the exercise" personally, but it sounds similar to some SGF papers that Gongwei Chen wrote. If you decide to rewrite the analysis in IML, there are a few articles about Markov chains and moving averages in IML that you might find helpful for writing efficient code:

 

View solution in original post


All Replies
Solution
‎04-03-2017 10:05 AM
SAS Super FREQ
Posts: 3,555

Re: forecasting delinquency

[ Edited ]

I cannot speak to "the exercise" personally, but it sounds similar to some SGF papers that Gongwei Chen wrote. If you decide to rewrite the analysis in IML, there are a few articles about Markov chains and moving averages in IML that you might find helpful for writing efficient code:

 

Valued Guide
Posts: 858

Re: forecasting delinquency

That's great, thanks Rick!

Super User
Posts: 18,589

Re: forecasting delinquency

No idea of the accuracy compared to a Markov model, but 2 stage regression are models are what I've seen. First stage calculates the probability of default and the second calculates the amount of the default assuming they're going to default.

 

 

☑ This topic is solved.

Need further help from the community? Please ask a new question.

Discussion stats
  • 3 replies
  • 147 views
  • 0 likes
  • 3 in conversation