|
Activity Number:
|
363
- Contributed Poster Presentations: IMS
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 5, 2020 : 10:00 AM to 2:00 PM
|
|
Sponsor:
|
IMS
|
|
Abstract #311096
|
|
|
Title:
|
Exact Simulation of Diffusions Revisited
|
|
Author(s):
|
Kumar Somnath* and Radu Herbei
|
|
Companies:
|
The Ohio State University and The Ohio State University
|
|
Keywords:
|
Exact simulation;
Diffusions;
rejection sampling;
Conditioned Brownian motion;
Bernoulli factory;
Brownian meander
|
|
Abstract:
|
We develop a new method for exact simulation of Stochastic Differential Equations(SDEs). We employ a rejection sampling method where first the sample paths are generated from a biased Brownian motion process and then they are accepted or rejected on the basis of the factor obtained via the Girsanov’s formula. The decision to accept or reject the proposed sample paths is done either via the use of a homogeneous Poisson process or via the Bernoulli factory. Our method requires an exact algorithm for simulation of maxima and time of maxima for a restricted Brownian meander and interpolation for Brownian motion restricted to an interval. These algorithms are themselves new additions to the encyclopedia of simulation methods for functionals of constrained Brownian motion and are used to provide more information about the proposed sample paths which eventually help in deciding whether to accept or reject them. Efficiency and example applications are demonstrated.
|
Authors who are presenting talks have a * after their name.
