Building models with SAS Enterprise Miner, SAS Factory Miner, SAS Visual Data Mining and Machine Learning or just with programming

Building Post-Disconnect Recovery Models

Reply
New Contributor
Posts: 4

Building Post-Disconnect Recovery Models

Our Analytics group has been tasked with building recovery models for our Credit & Collections department. I have tried researching the issues and pitfalls and I haven't been successful in finding anything specific to our task.
I am looking for helpful information on what kind of targets to use in modeling, modeling windows, performance windows, kinds of models used, etc.
I have heard of several different approaches like modeling for the optimal number of days to hold an account to maximize payment before sending it to an agency, or modeling for the highest net return, or even a two-stage model where you model on a binary target (pymt) and a continuous target (amount).
Without giving away company secrets, I am curious for any advice you can offer.

Thanks,
Andrew Christian
SAS Employee
Posts: 121

Re: Building Post-Disconnect Recovery Models

The best analytical approach to an analysis must take into account the availability of and quality of the data as well as the business objective.  It doesn't make sense to analyze data from the 'distant past' unless you believe the relationships will still hold for current data. In general, you would likely get a lot of different answers even with that detail available.   In many cases, the business needs will help to identify how large each window is.  

 

My assumption is that your recovery model would be tasked with identifying how to obtain the most revenue from people who still owed money after disonnecting their service.  These problems might be well-suited to a two-stage model -- one model which identifies whether you receive any money from someone after they disconnect, and the second model which predicts how much money you will be able to recover after they disconnect.  The product of these two models predictions generates an 'expected revenue per disconnect' which you could use to prioritize those accounts that will provide you the greatest revenue.  There is a TwoStage node that accepts both a binary (Yes/No) target and a value(e.g. revenue) target and allows you to build this model automatically.  You can also model the two targets separately.  

 

I hope this helps!

Doug 

 

Ask a Question
Discussion stats
  • 1 reply
  • 164 views
  • 0 likes
  • 2 in conversation