Nonasymptotic Estimation of Risk Measures Using Stochastic Gradient Langevin Dynamics
Jiarui Chu,Ludovic Tangpi
DOI: https://doi.org/10.1137/23m1552747
2024-06-09
SIAM Journal on Financial Mathematics
Abstract:SIAM Journal on Financial Mathematics, Volume 15, Issue 2, Page 503-536, June 2024. In this paper we will study the approximation of some law-invariant risk measures. As a starting point, we approximate the average value at risk using stochastic gradient Langevin dynamics, which can be seen as a variant of the stochastic gradient descent algorithm. Further, the Kusuoka spectral representation allows us to bootstrap the estimation of the average value at risk to extend the algorithm to general law-invariant risk measures. We will present both theoretical, nonasymptotic convergence rates of the approximation algorithm and numerical simulations.
social sciences, mathematical methods,mathematics, interdisciplinary applications,business, finance