Fair valuation under parameter uncertainty
Fair valuation of insurance liabilities is generally performed via a two-step approach, combining quadratic hedging with application of a risk measure on the residual liability, to obtain a cost-of-capital margin, see for instance [1]. This procedure requires the knowledge of the (joint) distribution of the liability and of the assets in the market, which is often not available in practice and estimated from historical data. Our goal is to find the best strategy/risk measure estimator that takes into account the riskiness arising from distribution uncertainty. In particular, focusing on the family of location-scale distributions, we consider elicitable risk measures and different (random) estimators, we study their properties and evaluate their accuracy. The approach is implemented using a fairly general neural network algorithm. Joint work with Salvatore Scognamoglio, University of Naples ‘Parthenope’ and Andreas Tsanakas, Bayes Business School, City University of London. References [1] J. Dhaene, B. Stassen, K. Barigou, D. Linders, and Z. Chen, Fair valuation of insurance liabilities: Merging actuarial judgement and market-consistency, Insurance: Mathematics and Economics, 76(2017), 14–27.
Area: CS14 - Stochastic models for risk and cooperation (Giacomo Scandolo and Alessandro Doldi)
Keywords: Fair valuation, elicitable risk measures, location-scale family, parameter uncertainty.
Please Login in order to download this file