Display options
Cite
Georgoulas A, Hillston J, Sanguinetti G. Unbiased Bayesian inference for population Markov jump processes via random truncations. Stat Comput. 2016;27(4):991-1002doi: 10.1007/s11222-016-9667-9.
Georgoulas, A., Hillston, J., & Sanguinetti, G. (2017). Unbiased Bayesian inference for population Markov jump processes via random truncations. Statistics and computing, 27(4), 991-1002. https://doi.org/10.1007/s11222-016-9667-9
Georgoulas, Anastasis, et al. "Unbiased Bayesian inference for population Markov jump processes via random truncations." Statistics and computing vol. 27,4 (2017): 991-1002. doi: https://doi.org/10.1007/s11222-016-9667-9
Georgoulas A, Hillston J, Sanguinetti G. Unbiased Bayesian inference for population Markov jump processes via random truncations. Stat Comput. 2017;27(4):991-1002. doi: 10.1007/s11222-016-9667-9. Epub 2016 Jun 02. PMID: 28690370; PMCID: PMC5477715.
Copy Download .nbib Format: NLM
Share it on

Stat Comput. 2017;27(4):991-1002. doi: 10.1007/s11222-016-9667-9. Epub 2016 Jun 02.

Unbiased Bayesian inference for population Markov jump processes via random truncations.

Statistics and computing

Anastasis Georgoulas, Jane Hillston, Guido Sanguinetti

Affiliations

  1. School of Informatics, University of Edinburgh, Edinburgh, UK.
  2. Synthetic and Systems Biology, University of Edinburgh, Edinburgh, UK.

PMID: 28690370 PMCID: PMC5477715 DOI: 10.1007/s11222-016-9667-9

Abstract

We consider continuous time Markovian processes where populations of individual agents interact stochastically according to kinetic rules. Despite the increasing prominence of such models in fields ranging from biology to smart cities, Bayesian inference for such systems remains challenging, as these are continuous time, discrete state systems with potentially infinite state-space. Here we propose a novel efficient algorithm for joint state/parameter posterior sampling in population Markov Jump processes. We introduce a class of pseudo-marginal sampling algorithms based on a random truncation method which enables a principled treatment of infinite state spaces. Extensive evaluation on a number of benchmark models shows that this approach achieves considerable savings compared to state of the art methods, retaining accuracy and fast convergence. We also present results on a synthetic biology data set showing the potential for practical usefulness of our work.

Keywords: Markov Chain Monte Carlo; Markov Jump Processes; Parameter estimation; Pseudo-marginal methods; Stochastic processes

References

  1. J Chem Phys. 2006 Jan 28;124(4):044104 - PubMed
  2. Nature. 2000 Jan 20;403(6767):339-42 - PubMed
  3. Nat Methods. 2014 Feb;11(2):197-202 - PubMed
  4. Biometrics. 2005 Sep;61(3):781-8 - PubMed
  5. Proc Natl Acad Sci U S A. 2006 May 30;103(22):8372-7 - PubMed
  6. Genetics. 2003 Jul;164(3):1139-60 - PubMed
  7. BMC Bioinformatics. 2009 Oct 19;10:343 - PubMed
  8. IEEE Trans Pattern Anal Mach Intell. 2014 Nov;36(11):2214-26 - PubMed
  9. Phys Rev Lett. 2009 Dec 4;103(23):230601 - PubMed
  10. Biometrics. 2014 Jun;70(2):457-66 - PubMed

Publication Types