Reversion Models: The 2 Best Ways to Calculate Historical Data

About the Podcast

In this episode, we’re delving into data selection, and why adjusting your data is even more important than forecasting. Gain insights into how to best utilize your data in your models. 

Listen On:

iTunesSpotifyApple PodcastRSSStitcher

Key Takeaways:

  • [07:47] If you are going to do a reversion model and you have a seven-year pool, and you are only going to focus 12 months that means that six years of your cash flow is going to be applied with a reversion percentage. Only one year of my cash flow is getting applied by a forecast.
  • [10:57] The first limitation is, do I have enough data to see how my loans went through prior economic cycles? History is the best indication of what is going to happen.  
  • [14:03] The vast majority of institutions are using DCF or PD models. Regardless of which model you use, you can apply any model to any pool, as long as you have the data.
  • [15:23] Aggregation is really important and when you combine that with the power of having two ways to look at your data you can really hone in on the risks. The secret is understanding the model.  

Presented By

Mike Umscheid
Mike Umscheid

President & CEO | ARCSys

Mike has been providing accounting, consulting and auditing services to financial institutions for over 30 years. Considered the “CECL Guru,” Mike was selected by the AICPA to create and deliver their 8-hour CPE course on CECL. He is a past member of the Auditing Standards Board and a published author on Accounting and Auditing for Financial Institutions. Mike has spoken at numerous AICPA conferences as well as other national and local financial institution associations.