Lévy process simulation by stochastic step functions

DOI10.1137/110851080MaRDI QIDQ2870641zbMATH OpenOpenAlexFDO

Authors Torquil Macdonald Sørensen, Fred Espen Benth

Publication date 21 January 2014

Published in SIAM Journal on Scientific Computing (Search for Journal in Brave)

Full work available at URL https://arxiv.org/abs/1110.2367

zbMATH Keywords

CGMYLévy processMarkov chainsMonte Carlo methodsNIGsimulation of probability distributions

Mathematics Subject Classification ID

Sample path properties (60G17) Processes with independent increments; Lévy processes (60G51) Computational methods in Markov chains (60J22) Time series, auto-correlation, regression, etc. in statistics (GARCH) (62M10) Monte Carlo methods (65C05)

Abstract: We study a Monte Carlo algorithm for simulation of probability distributions based on stochastic step functions, and compare to the traditional Metropolis/Hastings method. Unlike the latter, the step function algorithm can produce an uncorrelated Markov chain. We apply this method to the simulation of Levy processes, for which simulation of uncorrelated jumps are essential. We perform numerical tests consisting of simulation from probability distributions, as well as simulation of Levy process paths. The Levy processes include a jump-diffusion with a Gaussian Levy measure, as well as jump-diffusion approximations of the infinite activity NIG and CGMY processes. To increase efficiency of the step function method, and to decrease correlations in the Metropolis/Hastings method, we introduce adaptive hybrid algorithms which employ uncorrelated draws from an adaptive discrete distribution defined on a space of subdivisions of the Levy measure space. The nonzero correlations in Metropolis/Hastings simulations result in heavy tails for the Levy process distribution at any fixed time. This problem is eliminated in the step function approach. In each case of the Gaussian, NIG and CGMY processes, we compare the distribution at t=1 with exact results and note the superiority of the step function approach.

Review text

The authors ``study a Monte Carlo algorithm for simulation of probability distributions based on stochastic step functions, and compare it to the traditional Metropolis/Hastings method. The latter method produces subsequent values which are not independent, due to the acceptance/rejection nature of this algorithm. The advantage of the new method is that an uncorrelated Markov chain can be obtained. The authors ``apply this method to the simulation of Lévy processes, for which simulation of uncorrelated jumps are essential. Furthermore, they ``perform numerical tests consisting of simulation from probability distributions, as well as simulation of Lévy process paths. The processes under consideration ``include a jump-diffusion with a Gaussian Lévy measure, as well as jump-diffusion approximations of the infinite activity NIG and CGMY processes. To increase efficiency of the step function method, and to decrease correlations in the Metropolis/Hastings method, (they) introduce adaptive hybrid algorithms which employ uncorrelated draws from an adaptive discrete distribution defined on a space of subdivisions of the Lévy measure space.

Recommendations

• New methods of simulating Lévy processes
• Stratified sampling and quasi-Monte Carlo simulation of Lévy processes
• A Wiener-Hopf Monte Carlo simulation technique for Lévy processes
• Multilevel Monte Carlo simulation for Lévy processes based on the Wiener-Hopf factorisation
• Applying the Wiener-Hopf Monte Carlo simulation technique for Lévy processes to path functionals

Cited in

(14)

• New methods of simulating Lévy processes
• Simulation of Student-Lévy processes using series representations
• Multilevel Monte Carlo simulation for Lévy processes based on the Wiener-Hopf factorisation
• Simulation of stochastic integrals with respect to Lévy processes of type G.
• Simulation of the drawdown and its duration in Lévy models via stick-breaking Gaussian approximation
• Stratified sampling and quasi-Monte Carlo simulation of Lévy processes
• Monte Carlo methods for radiative transfer with singular kernels
• Simulating Lévy processes from their characteristic functions and financial applications
• Isotropic random fields with infinitely divisible marginal distributions
• Applying the Wiener-Hopf Monte Carlo simulation technique for Lévy processes to path functionals
• Vector Stochastic Processes with Pólya‐Type Correlation Structure
• Point process simulation of generalised hyperbolic Lévy processes
• Simulating stochastic diffusion processes and processes with ``market time
• Exact simulation of a truncated Lévy subordinator

This page was built for publication: Lévy process simulation by stochastic step functions

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2870641)