Home > Research > Publications & Outputs > Discussion on the paper of 'Particle Markov cha...
View graph of relations

Discussion on the paper of 'Particle Markov chain Monte Carlo methods' by Christophe Andrieu, Arnaud Doucet, and Roman Holenstein

Research output: Contribution to Journal/MagazineComment/debatepeer-review

Published

Standard

Discussion on the paper of 'Particle Markov chain Monte Carlo methods' by Christophe Andrieu, Arnaud Doucet, and Roman Holenstein. / Gonzalez Belmonte, Miguel Angel; Papaspiliopoulos, Omiros.
In: Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 72, No. 3, 2010, p. 308-310.

Research output: Contribution to Journal/MagazineComment/debatepeer-review

Harvard

APA

Vancouver

Gonzalez Belmonte MA, Papaspiliopoulos O. Discussion on the paper of 'Particle Markov chain Monte Carlo methods' by Christophe Andrieu, Arnaud Doucet, and Roman Holenstein. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2010;72(3):308-310. doi: 10.1111/j.1467-9868.2009.00736.x

Author

Bibtex

@article{75129ff41f524893bf3215db99244938,
title = "Discussion on the paper of 'Particle Markov chain Monte Carlo methods' by Christophe Andrieu, Arnaud Doucet, and Roman Holenstein",
abstract = "We congratulate the authors for a remarkable paper, which addresses a problem of fundamental practical importance: parameter estimation in state space models by using sequential Monte Carlo (SMC) algorithms. In Belmonte et al. (2008) we fit duration state space models to high frequency transaction data and we require a computational methodology that can handle efficiently time series of length T =O.104–105/. We have experimented with particle Markov chain Monte Carlo (PMCMC) methods and with the smooth particle filter (SPF) of Pitt (2002). The latter is also based on the use of SMC algorithms to derive maximum likelihood parameter estimates; it is, however, limited to scalar signals. Therefore, in the context of duration modelling this limitation rules out multifactor or multi-dimensional models, and we believe that PMCMC methods can be very useful in such cases.",
keywords = "Bayesian inference, Markov chain Monte Carlo methods, Sequential Monte Carlo methods, State space models",
author = "{Gonzalez Belmonte}, {Miguel Angel} and Omiros Papaspiliopoulos",
note = "The paper was read before The Royal Statistical Society at a meeting organized by the Research Section on Wednesday, October 14th, 2009.",
year = "2010",
doi = "10.1111/j.1467-9868.2009.00736.x",
language = "English",
volume = "72",
pages = "308--310",
journal = "Journal of the Royal Statistical Society: Series B (Statistical Methodology)",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "3",

}

RIS

TY - JOUR

T1 - Discussion on the paper of 'Particle Markov chain Monte Carlo methods' by Christophe Andrieu, Arnaud Doucet, and Roman Holenstein

AU - Gonzalez Belmonte, Miguel Angel

AU - Papaspiliopoulos, Omiros

N1 - The paper was read before The Royal Statistical Society at a meeting organized by the Research Section on Wednesday, October 14th, 2009.

PY - 2010

Y1 - 2010

N2 - We congratulate the authors for a remarkable paper, which addresses a problem of fundamental practical importance: parameter estimation in state space models by using sequential Monte Carlo (SMC) algorithms. In Belmonte et al. (2008) we fit duration state space models to high frequency transaction data and we require a computational methodology that can handle efficiently time series of length T =O.104–105/. We have experimented with particle Markov chain Monte Carlo (PMCMC) methods and with the smooth particle filter (SPF) of Pitt (2002). The latter is also based on the use of SMC algorithms to derive maximum likelihood parameter estimates; it is, however, limited to scalar signals. Therefore, in the context of duration modelling this limitation rules out multifactor or multi-dimensional models, and we believe that PMCMC methods can be very useful in such cases.

AB - We congratulate the authors for a remarkable paper, which addresses a problem of fundamental practical importance: parameter estimation in state space models by using sequential Monte Carlo (SMC) algorithms. In Belmonte et al. (2008) we fit duration state space models to high frequency transaction data and we require a computational methodology that can handle efficiently time series of length T =O.104–105/. We have experimented with particle Markov chain Monte Carlo (PMCMC) methods and with the smooth particle filter (SPF) of Pitt (2002). The latter is also based on the use of SMC algorithms to derive maximum likelihood parameter estimates; it is, however, limited to scalar signals. Therefore, in the context of duration modelling this limitation rules out multifactor or multi-dimensional models, and we believe that PMCMC methods can be very useful in such cases.

KW - Bayesian inference

KW - Markov chain Monte Carlo methods

KW - Sequential Monte Carlo methods

KW - State space models

U2 - 10.1111/j.1467-9868.2009.00736.x

DO - 10.1111/j.1467-9868.2009.00736.x

M3 - Comment/debate

VL - 72

SP - 308

EP - 310

JO - Journal of the Royal Statistical Society: Series B (Statistical Methodology)

JF - Journal of the Royal Statistical Society: Series B (Statistical Methodology)

SN - 1369-7412

IS - 3

ER -