Multilevel Monte Carlo metamodeling

Author

Rosenbaum, Imry ; Staum, Jeremy

Author_Institution

Dept. of Ind. Eng. & Manage. Sci., Northwestern Univ., Evanston, IL, USA

fYear

2013

fDate

8-11 Dec. 2013

Firstpage

509

Lastpage

520

Abstract

Multilevel Monte Carlo (MLMC) methods have been used by the information-based complexity community in order to improve the computational efficiency of parametric integration. We extend this approach by relaxing the assumptions on differentiability of the simulation output. Relaxing the assumption on the differentiability of the simulation output makes the MLMC method more widely applicable to stochastic simulation metamodeling problems in industrial engineering. The proposed scheme uses a sequential experiment design which allocates effort unevenly among design points in order to increase its efficiency. The procedure´s efficiency is tested on an example of option pricing in the Black-Scholes model.

Keywords

Monte Carlo methods; computational complexity; design of experiments; industrial engineering; modelling; Black-Scholes model; MLMC methods; design points; differentiability; industrial engineering; information-based complexity; multilevel Monte Carlo metamodeling; option pricing; parametric integration; sequential experiment design; simulation output; stochastic simulation metamodeling problems; Computational modeling; Function approximation; Metamodeling; Monte Carlo methods; Random variables; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Simulation Conference (WSC), 2013 Winter

Conference_Location

Washington, DC

Print_ISBN

978-1-4799-2077-8

Type

conf

DOI

10.1109/WSC.2013.6721446

Filename

6721446