Manual técnico de IBNR, reserving y modelación actuarial para seguros de salud, con aplicaciones al sistema de salud colombiano.
-
Updated
Jul 18, 2026 - Python
Manual técnico de IBNR, reserving y modelación actuarial para seguros de salud, con aplicaciones al sistema de salud colombiano.
Chain-ladder claims reserving (development factors, ultimates, IBNR) with backtesting.
IBNR reserve estimation — Bootstrap Chain-Ladder | 10,000 Monte-Carlo simulations | OSFI MCT 99.5% VaR | CIA Standards | Ontario Auto BI
ML-EM nowcasting for claims reporting delays — joint Poisson-Multinomial EM, XGBoost/GLM M-step, exposure offset, bootstrap CIs
RateCraft: health-plan actuarial pricing and IBNR reserving engine. Chain-ladder + Bornhuetter-Ferguson validated against the published Taylor-Ashe/Mack triangle, credibility-blended rate development, and cell-exact Excel exhibits. 77 tests, 100% coverage.
Claims reserving: chain ladder, Bornhuetter-Ferguson, Mack, and the ODP bootstrap of the predictive reserve distribution.
General insurance loss reserving in R: chain ladder, Bornhuetter-Ferguson, Cape Cod and Benktander, with Mack standard error, an ODP bootstrap, and a risk margin at the 75% sufficiency level under the Bank Negara Malaysia RBC framework.
Temporal cross-validation for insurance pricing models. Walk-forward splits respecting policy/accident year structure and IBNR development buffers.
Chain-ladder IBNR reserving in base R. Deterministic estimate plus bootstrap risk capital (VaR/RCS), from scratch, no dependencies.
Add a description, image, and links to the ibnr topic page so that developers can more easily learn about it.
To associate your repository with the ibnr topic, visit your repo's landing page and select "manage topics."