ODP Bootstrap Reserve Range
Simulate the full distribution of the chain-ladder reserve using an over-dispersed Poisson bootstrap on Pearson residuals, with percentile ranges and a histogram.
Loaded: State Farm Mut Grp — Workers' Compensation (paid, CAS Loss Reserve DB)
| Origin \ Dev | 1 | 2 | 3 | 4 | 5 | 6 | 7 | 8 | 9 | 10 |
|---|---|---|---|---|---|---|---|---|---|---|
| AY 1 | ||||||||||
| AY 2 | ||||||||||
| AY 3 | ||||||||||
| AY 4 | ||||||||||
| AY 5 | ||||||||||
| AY 6 | ||||||||||
| AY 7 | ||||||||||
| AY 8 | ||||||||||
| AY 9 | ||||||||||
| AY 10 |
Assumptions & limitations
- Incremental losses follow an over-dispersed Poisson process; Pearson residuals are treated as i.i.d. and exchangeable across the triangle.
- Each simulation re-fits volume-weighted chain-ladder factors on the resampled pseudo-triangle, then adds process variance around the resulting IBNR.
- Results are seeded (mulberry32 PRNG) for reproducibility; re-running with the same seed and N reproduces the same distribution.