SLIDE 1
From Survival Analysis to Projecting Disability Payments
Benjamin Abt & Simon Adamov January 30, 2020
SLIDE 2 Disclaimer
The following presentation is for general information, education and discussion purposes only. It may not be reproduced or disseminated in any form, without the prior written permission of PartnerRe. Views or opinions expressed, whether oral or in writing, do not necessarily reflect those of PartnerRe, nor do they constitute legal or professional advice. PartnerRe accepts no liability as a result of any reliance you may have placed or action taken based upon the information outlined in this presentation.
SLIDE 3
Introduction
SLIDE 4
Group disability insurance
SLIDE 5
Group disability insurance
SLIDE 6
Group disability insurance
SLIDE 7
Group disability insurance
SLIDE 8
Group disability insurance
SLIDE 9
Group disability insurance
SLIDE 10
Group disability insurance
SLIDE 11
Disability insurance: Classic claim unwinding
SLIDE 12
Disability insurance: Classic claim unwinding
SLIDE 13
Disability insurance: Classic claim unwinding
SLIDE 14
Disability insurance: Classic claim unwinding
SLIDE 15
Disability insurance: Classic claim unwinding
SLIDE 16
Disability insurance: Classic claim unwinding
SLIDE 17
Disability insurance: Classic claim unwinding
SLIDE 18
Disability insurance: Classic claim unwinding
SLIDE 19
Disability insurance: Classic claim unwinding
SLIDE 20
Disability insurance: Classic claim unwinding
SLIDE 21
Disability insurance: Classic claim unwinding
SLIDE 22
Disability insurance: Open claim and censoring
SLIDE 23
Disability insurance: Open claim and censoring
SLIDE 24
Disability insurance: Open claim and censoring
SLIDE 25
Disability insurance: Open claim and censoring
SLIDE 26
Disability insurance: Open claim and censoring
SLIDE 27 A termination study evolving into a reserving tool
Project emerged from a strong client-PartnerRe relationship. Provided us with claims data to perform multivariate analysis on claims duration. Multiple iterations on the course of 4 years. Developed into a tool for projecting the cost for the (re)insurer of such disability claims.
SLIDE 28
Data
SLIDE 29
SLIDE 30
Modeling
SLIDE 31 fit_km <- survival::survfit(Surv(openTime, closed) ~ 1, data = claims_new)
SLIDE 32
SLIDE 33 survival::coxph party::cforest Advantages Experience Few Assumptions Transparent Individual Curves Computing Time Good Performance Disadvantages PH Assumption Hard to Interpret Uncertain Event Times Data Hungry
tree <- party::ctree(Surv(openTime, closed) ~ ..., data = train_forest)
Comparison of CoxPH and cForest
SLIDE 34 pec <- pec::pec(list("cox" = fit_coxph, "cforest" = fit_forest_pec), data = claims, formula = form, splitMethod = "cv10") zph <- survival::cox.zph(fit_coxph, transform = "km") gg_zph <- survminer::ggcoxzph(zph)
Evaluation Metrics
SLIDE 35 200 400 600 800 0% 25% 50% 75% 100%
KM Estimator KM ConfInt CoxPH CoxPH ConfInt cForest
Comparison of Survival Curves and Confidence Intervals Time [Days] Survival Probability
Model-Type
SLIDE 36
Packaging
SLIDE 37
Projecting disability probabilities
SLIDE 38
Projecting disability probabilities
SLIDE 39 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 40 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 41 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 42 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 43 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 44 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 45 A two packages set-up
Models Package Reserve Calculation Package
Survival Analysis Models Data handling process Model fitting code Functions to access models Functions to compute reserves ReadMe, vignettes, tests & pkgDown Tails ReadMe, vignettes, tests & pkgDown Data mask Economic assumptions
SLIDE 46
Conclusion
SLIDE 47 From survival analysis to projecting disability payments
Yearly evaluation of released R tools and packages deepens our understanding of survival analysis. Successful textbook case of wrapping-up an analysis into R packages. R and its packages enables the construct of production tools in business environments.
