Bailey and Simon Flashcards
Experience Mod
Premium adjustements for relative frequency calculation
Maldistribution
Exposure correlation (i.e. between territory and merit)
Choosing premium base for frequency
Mod formula
Bailey and Simon assumptions for calculating R
- Class total claim frequency to ECY is the same each year
- Claim counts are Poisson distributed
R formula
Experience rating credibility dependencies
Volume of data
Variance of loss experience within a class (Experience rating distinguishes the individual risk from the class average risk)
2-year, 3-year credibilities relative to 1-year
The closer the 2- and 3- year relativities are to 2.0 and 3.0, respectively, the more stable the book:
1) Risks entering/exiting portfolio
2) Risk characteristics changing over time
3 conclusions of Bailey and Simon paper
Buhlmann credibility formula
Hazam conclusions
- Credibility increases in proportion to years only for low credibility values
- Simple Buhlmann credibility shows that double or triple the data doesn’t result in exactly double or triple credibility (since k does not change for different samples of n sizes taken from the X variable)
Z = n / (n + k)
K, Buhlmann credibility
K = EVPV / VHM
= E[Var(X¦µ)] / Var(E[X¦µ])
Why individual risk experience is more credible when there is m ore variance in loss experience
Experience rating distinguishes the individual risk from the class average risk