Automating Reserving for optimal model selection.

Automating Reserving for optimal model selection.


Thanks! Share it with your friends!


You disliked this video. Thanks for the feedback!

Sorry, only registred users can create playlists.


There are a wide range of Reserving methods out there, but due to time constraints and prior preferences, actuaries very often confine themselves to a single or minimal number of methods in their work. 

As certain trends very often invalidate the underlying assumptions and historic statistical development patterns, there is a risk that reserves may be set inappropriately.  Certain methods perform better under different circumstances, and the actuary would need indicators that highlight when to switch between methods in order to avoid pitfalls.

Even without distortions in development data, there is significant uncertainty in setting reserve estimates due to the complexity and randomness of the claims process.

Using automation, back-testing and visual dashboards, it is possible to set up rules to highlight which methods are performing the best in terms of accuracy and where previous choice of method becomes invalidated.

Structure of presentation

  1. Provide overview on a wide range of reserving methods out there…there is life beyond chain ladder and BF.
  2. Present visual dashboards and metrics that can be used to test methods and selections.
  3. Provide overview of approach to automated model selection and automated prior assumption setting.

Post your comment

Sign in or sign up to post comments.
Be the first to comment