Casualty Actuarial Society

2013 Casualty Loss Reserve Seminar

VR-8
Stochastic Loss Reserving with Bayesian MCMC Models

Tuesday, September 17, 2013: 10:30 a.m.
Wellesley Room (Boston Marriott Copley Place)
The emergence of Bayesian Markov Chain Monte-Carlo (MCMC) models has provided actuaries with an unprecedented flexibility in stochastic model development.  Another recent development has been the posting of a database on the CAS website that consists of hundreds of loss development triangles with outcomes.  This session begins by first testing the performance of the Mack model on incurred data, and the Bootstrap Overdispersed Poisson model on paid data.  It then will discuss features of Bayesian MCMC models that improve the performance over the above models.  These features include (1) allowing for correlation between accident years; (2) introducing a skewed distribution defined over the entire real line to deal with negative incremental paid data; and (3) allowing for a payment year trend on paid data.
Moderator: Glenn Meyers, Retired, None

Speakers

Presentation 1
Glenn Meyers, Retired, None
Handouts
  • CLRS 2013 SLRMCMC - Meyers.pdf (1.2 MB)
  • See more of: Conference Program