Exact simulation of diffusions
classification
🧮 math.PR
keywords
exactpathalgorithmapplicableclasscollectioncompleteddescribe
read the original abstract
We describe a new, surprisingly simple algorithm, that simulates exact sample paths of a class of stochastic differential equations. It involves rejection sampling and, when applicable, returns the location of the path at a random collection of time instances. The path can then be completed without further reference to the dynamics of the target process.
This paper has not been read by Pith yet.
Forward citations
Cited by 1 Pith paper
-
Quasi-Monte Carlo for SDE Simulation: Error Analysis and Dimensionality Reduction
QMC applied to Euler-Maruyama yields faster sampling-error decay than Monte Carlo, and the new MSTG method based on exact simulation achieves super-exponential truncation-error decay that sharply reduces integration d...
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.