Good Deal Hedging and Valuation under Combined Uncertainty about Drift and Volatility

Dirk Becherer,Klebert Kentia
DOI: https://doi.org/10.48550/arXiv.1704.02505
2017-04-08
Abstract:We study robust notions of good-deal hedging and valuation under combined uncertainty about the drifts and volatilities of asset prices. Good-deal bounds are determined by a subset of risk-neutral pricing measures such that not only opportunities for arbitrage are excluded but also deals that are too good, by restricting instantaneous Sharpe ratios. A non-dominated multiple priors approach to model uncertainty (ambiguity) leads to worst-case good-deal bounds. Corresponding hedging strategies arise as minimizers of a suitable coherent risk measure. Good-deal bounds and hedges for measurable claims are characterized by solutions to second-order backward stochastic differential equations whose generators are non-convex in the volatility. These hedging strategies are robust with respect to uncertainty in the sense that their tracking errors satisfy a supermartingale property under all a-priori valuation measures, uniformly over all priors.
Mathematical Finance,Optimization and Control,Probability,Portfolio Management
What problem does this paper attempt to address?