Leveraging predictive models: The right approach for your institution?

By: Lauren Smith

Because predictive models will have a direct impact on an institution’s ability to support the CECL estimate, institutions should thoroughly evaluate the costs and benefits of using a modeled and non-modeled approach. Predictive models require significant investment, including upfront development and validation as well as ongoing maintenance. However, the benefits of leveraging predictive models include a streamlined reserving process and greater transparency into a potentially volatile estimate.

The whitepaper, “CECL Preparation: An Introduction to Models for Accountants and Other Non-Modelers”, written by John Lankenau, expands on the concepts recently discussed during the webinar of the same title, hosted by SS&C Primatics. The whitepaper presents a decision-making framework for evaluating the use of predictive models for the CECL estimate.

The whitepaper includes the following:

  • The definition of a model, along with the key concepts and characteristics of good models
  • Estimating the impact of reasonable and supportable forecasts using predictive models compared to a non-modeled approach
  • An illustrative example of incorporating reasonable and supportable forecasts using a non-modeled approach
  • Costs and benefits of the reserving lifecycle using a modeled and non-modeled approach

Predictive models can be used to support the CECL estimate and streamline the reserving process. To learn more about the tools your firm needs to successfully transition to CECL, read “CECL Preparation: An Introduction to Models for Accountants and Other Non-Modelers” or contact us for a demo at solution@sscinc.com or 800-234-0556.