Sampling from mixture distributions based on regime-switching diffusions

Tretyakov, M. V.

Abstract

It is proposed to use stochastic differential equations with state-dependent switching 4 rates (SDEwS) for sampling from finite mixture distributions. An Euler scheme with constant time 5 step for SDEwS is considered. It is shown that the scheme converges with order one in weak sense 6 and also in the ergodic limit. Numerical experiments illustrate the use of SDEwS for sampling from 7 mixture distributions and confirm the theoretical results. 8 Key words. Stochastic differential equations with switching, hybrid switching diffusions, finite 9 mixture distributions, ergodic limits, sampling, weak numerical schemes.

Citation

Tretyakov, M. V. (in press). Sampling from mixture distributions based on regime-switching diffusions. SIAM Journal on Scientific Computing,

• Journal Article Type: Article
• Acceptance Date: Jan 29, 2025
• Deposit Date: Jan 31, 2025
• Publicly Available Date: Jan 31, 2025
• Journal: SIAM Journal on Scientific Computing
• Print ISSN: 1064-8275
• Electronic ISSN: 1095-7197
• Publisher: Society for Industrial and Applied Mathematics
• Peer Reviewed: Peer Reviewed
• Keywords: Stochastic differential equations with switching, hybrid switching diffusions, finite mixture distributions, ergodic limits, sampling, weak numerical schemes.
• Public URL: https://nottingham-repository.worktribe.com/output/44692341
• Publisher URL: https://epubs.siam.org/journal/sjoce3