Backend
An explicit domain-preserving projection scheme for scalar stochastic differential equations
arXiv cs.NA0 views
arXiv:2611.23522v1 Announce Type: new Abstract: We develop an explicit domain-preserving projection scheme for scalar stochastic differential equations whose solutions evolve in a prescribed interval. A Lamperti-type transformation converts the original equation into one with additive noise, whose drift may be singular near finite boundaries or grow polynomially at infinity. The proposed domain-preserving projected Euler-Maruyama (DPPEM) scheme combines an explicit Euler-Maruyama step with projection onto a step-size-dependent interior interval. Under one-sided Lipschitz, boundary, and weighted growth assumptions, with projection parameters chosen according to the coefficient growth and available moment orders, we establish strong order one convergence for admissible moment orders. The analysis combines local consistency and stability estimates with moment and boundary negative moment bounds for the exact solution. Numerical moment bounds follow from the convergence estimate, while numerical negative moment bounds are obtained under additional conditions. Estimates for the inverse Lamperti map yield strong order one convergence for the original equations in the CIR, Heston 3/2, CEV, Ait-Sahalia, and bounded interest-rate models under the stated model conditions. Numerical experiments illustrate the predicted convergence rates and compare computational costs with the Lamperti backward Euler-Maruyama scheme.
Source:https://arxiv.org/abs/2611.23522PreviousOptimal Control of RKHS Diffeomorphic Flows for Classification and Generative ModelingNextModeling fluid--solid interaction in three-dimensional granular systems: Two-way coupling of SPH and LS-DEM
Related posts
Shipping an LLM feature to production: a 6-step checklist from real client projects
0dev.to
Why I Built an Ad-Free, 4K Video Downloader in Next.js 14 & Python (And Killed Popups Forever)
0dev.to
Three Hundred of Our Trade Accounts Belonged to Companies That No Longer Existed
0dev.to
Title40-AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
0dev.to
Title50-AAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAAA
0dev.to

Practical: a webcam physiology lab that grades itself against a wrist count
0dev.to