EURO-Online login
- New to EURO? Create an account
- I forgot my username and/or my password.
- Help with cookies
(important for IE8 users)
3613. Value-at-Risk estimation in nested simulations
Invited abstract in session MB-57: Decision making in Insurance and Pensions, stream Modern Decision Making in Finance and Insurance.
Monday, 10:30-12:00Room: S06 (building: 101)
Authors (first author is the speaker)
| 1. | Alexander Strack
|
Abstract
Since the ratification of Solvency II, the determination of the Value-at-Risk in the insurance industry has gained renewed importance. We examine the convergence of estimators of the Value-at-Risk or, more specifically, of quantiles within the nested Monte Carlo framework as they occur in the internal models of (typical life) insurance companies. Due to the numerous model uncertainties, robust methods and model-free results are of particular importance. To this end, we provide sharp results regarding the deviation of the quantiles of a perturbed random variable to the quantile of the undisturbed random variable depending on the perturbation, while reducing the assumptions to a minimum without sacrificing the sharpness of the results. These results are combined with classical results from estimation theory to derive almost-sure convergence rates for the Value-at-Risk estimator under rather weak assumptions.
Keywords
- Risk Analysis and Management
- Simulation
- Finance and Banking
Status: accepted
Back to the list of papers