Home > Research > Publications & Outputs > Efficient sampling of conditioned Markov jump p...

Electronic data

  • 1809.07139

    Accepted author manuscript, 424 KB, PDF document

Links

Text available via DOI:

View graph of relations

Efficient sampling of conditioned Markov jump processes

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Efficient sampling of conditioned Markov jump processes. / Golightly, Andrew; Sherlock, Christopher Gerrard.
In: Statistics and Computing, 13.02.2019.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

APA

Golightly, A., & Sherlock, C. G. (2019). Efficient sampling of conditioned Markov jump processes. Statistics and Computing. Advance online publication. https://doi.org/10.1007/s11222-019-09861-5

Vancouver

Golightly A, Sherlock CG. Efficient sampling of conditioned Markov jump processes. Statistics and Computing. 2019 Feb 13. Epub 2019 Feb 13. doi: 10.1007/s11222-019-09861-5

Author

Golightly, Andrew ; Sherlock, Christopher Gerrard. / Efficient sampling of conditioned Markov jump processes. In: Statistics and Computing. 2019.

Bibtex

@article{66791e0b4a0a4ac58074505b5600a3fc,
title = "Efficient sampling of conditioned Markov jump processes",
abstract = "We consider the task of generating draws from a Markov jump process (MJP) between two time-points at which the process is known. Resulting draws are typically termed bridges and the generation of such bridges plays a key role in simulation-based inference algorithms for MJPs. The problem is challenging due to the intractability of the conditioned process, necessitating the use of computationally intensive methods such as weighted resampling or Markov chain Monte Carlo. An efficient implementation of such schemes requires an approximation of the intractable conditioned hazard/propensity function that is both cheap and accurate. In this paper, we review some existing approaches to this problem before outlining our novel contribution. Essentially, we leverage the tractability of a Gaussian approximation of the MJP and suggest a computationally efficient implementation of the resulting conditioned hazard approximation. We compare and contrast our approach with existing methods using three examples.",
keywords = "Markov jump process, Conditioned hazard, Chemical Langevin equation, Linear noise approximation",
author = "Andrew Golightly and Sherlock, {Christopher Gerrard}",
year = "2019",
month = feb,
day = "13",
doi = "10.1007/s11222-019-09861-5",
language = "English",
journal = "Statistics and Computing",
issn = "0960-3174",
publisher = "Springer Netherlands",

}

RIS

TY - JOUR

T1 - Efficient sampling of conditioned Markov jump processes

AU - Golightly, Andrew

AU - Sherlock, Christopher Gerrard

PY - 2019/2/13

Y1 - 2019/2/13

N2 - We consider the task of generating draws from a Markov jump process (MJP) between two time-points at which the process is known. Resulting draws are typically termed bridges and the generation of such bridges plays a key role in simulation-based inference algorithms for MJPs. The problem is challenging due to the intractability of the conditioned process, necessitating the use of computationally intensive methods such as weighted resampling or Markov chain Monte Carlo. An efficient implementation of such schemes requires an approximation of the intractable conditioned hazard/propensity function that is both cheap and accurate. In this paper, we review some existing approaches to this problem before outlining our novel contribution. Essentially, we leverage the tractability of a Gaussian approximation of the MJP and suggest a computationally efficient implementation of the resulting conditioned hazard approximation. We compare and contrast our approach with existing methods using three examples.

AB - We consider the task of generating draws from a Markov jump process (MJP) between two time-points at which the process is known. Resulting draws are typically termed bridges and the generation of such bridges plays a key role in simulation-based inference algorithms for MJPs. The problem is challenging due to the intractability of the conditioned process, necessitating the use of computationally intensive methods such as weighted resampling or Markov chain Monte Carlo. An efficient implementation of such schemes requires an approximation of the intractable conditioned hazard/propensity function that is both cheap and accurate. In this paper, we review some existing approaches to this problem before outlining our novel contribution. Essentially, we leverage the tractability of a Gaussian approximation of the MJP and suggest a computationally efficient implementation of the resulting conditioned hazard approximation. We compare and contrast our approach with existing methods using three examples.

KW - Markov jump process

KW - Conditioned hazard

KW - Chemical Langevin equation

KW - Linear noise approximation

U2 - 10.1007/s11222-019-09861-5

DO - 10.1007/s11222-019-09861-5

M3 - Journal article

JO - Statistics and Computing

JF - Statistics and Computing

SN - 0960-3174

ER -