Sampling Of One-Dimensional Probability Measures In The Convex Order And Computation Of Robust Option Price Bounds
AURÉLIEN ALFONSI,JACOPO CORBETTA,BENJAMIN JOURDAIN,Aurelien Alfonsi,Jacopo Corbetta,Benjamin Jourdain
DOI: https://doi.org/10.1142/S021902491950002X
2019-10-05
International Journal of Theoretical and Applied Finance
Abstract:International Journal of Theoretical and Applied Finance, Ahead of Print. For [math] and [math] two probability measures on the real line such that [math] is smaller than [math] in the convex order, this property is in general not preserved at the level of the empirical measures [math] and [math], where [math] (resp., [math]) are independent and identically distributed according to [math] (resp., [math]). We investigate modifications of [math] (resp., [math]) smaller than [math] (resp., greater than [math]) in the convex order and weakly converging to [math] (resp., [math]) as [math]. According to Kertz & Rösler (1992), the set of probability measures on the real line with a finite first order moment is a complete lattice for the increasing and the decreasing convex orders. For [math] and [math] in this set, this enables us to define a probability measure [math] (resp., [math]) greater than [math] (resp., smaller than [math]) in the convex order. We give efficient algorithms permitting to compute [math] and [math] (and therefore [math] and [math]) when [math] and [math] have finite supports. Last, we illustrate by numerical experiments the resulting sampling methods that preserve the convex order and their application to approximate martingale optimal transport problems and in particular to calculate robust option price bounds.