Home > Research > Publications & Outputs > Online Inference for Multiple Changepoint Probl...

Research

Electronic data

Links

Text available via DOI:

Keywords

View graph of relations

Online Inference for Multiple Changepoint Problems.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Published

Standard

Online Inference for Multiple Changepoint Problems. / Fearnhead, P; Liu, Z.
In: Journal of the Royal Statistical Society: Series B (Statistical Methodology), Vol. 69, No. 4, 2007, p. 589-605.

Research output: Contribution to Journal/Magazine › Journal article › peer-review

Harvard

Fearnhead, P & Liu, Z 2007, 'Online Inference for Multiple Changepoint Problems.', Journal of the Royal Statistical Society: Series B (Statistical Methodology), vol. 69, no. 4, pp. 589-605. https://doi.org/10.1111/j.1467-9868.2007.00601.x

APA

Fearnhead, P., & Liu, Z. (2007). Online Inference for Multiple Changepoint Problems. Journal of the Royal Statistical Society: Series B (Statistical Methodology), 69(4), 589-605. https://doi.org/10.1111/j.1467-9868.2007.00601.x

Vancouver

Fearnhead P, Liu Z. Online Inference for Multiple Changepoint Problems. Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2007;69(4):589-605. doi: 10.1111/j.1467-9868.2007.00601.x

Author

Fearnhead, P ; Liu, Z. / Online Inference for Multiple Changepoint Problems. In: Journal of the Royal Statistical Society: Series B (Statistical Methodology). 2007 ; Vol. 69, No. 4. pp. 589-605.

Bibtex

@article{42009dd8d3674354ae82383c6eba7c97,
title = "Online Inference for Multiple Changepoint Problems.",
abstract = "We propose an on-line algorithm for exact filtering of multiple changepoint problems. This algorithm enables simulation from the true joint posterior distribution of the number and position of the changepoints for a class of changepoint models. The computational cost of this exact algorithm is quadratic in the number of observations. We further show how resampling ideas from particle filters can be used to reduce the computational cost to linear in the number of observations, at the expense of introducing small errors; and propose two new, optimum resampling algorithms for this problem. One, a version of rejection control, allows the particle filter to automatically choose the number of particles required at each time-step. The new resampling algorithms substantially out-perform standard resampling algorithms on examples we consider; and we demonstrate how the resulting particle filter is practicable for segmentation of human GC content.",
keywords = "Direct simulation, Isochores, Rejection control, Sequential Monte Carlo, Stratified sampling, Particle Filtering",
author = "P Fearnhead and Z Liu",
year = "2007",
doi = "10.1111/j.1467-9868.2007.00601.x",
language = "English",
volume = "69",
pages = "589--605",
journal = "Journal of the Royal Statistical Society: Series B (Statistical Methodology)",
issn = "1369-7412",
publisher = "Wiley-Blackwell",
number = "4",

}

RIS

TY - JOUR

T1 - Online Inference for Multiple Changepoint Problems.

AU - Fearnhead, P

AU - Liu, Z

PY - 2007

Y1 - 2007

N2 - We propose an on-line algorithm for exact filtering of multiple changepoint problems. This algorithm enables simulation from the true joint posterior distribution of the number and position of the changepoints for a class of changepoint models. The computational cost of this exact algorithm is quadratic in the number of observations. We further show how resampling ideas from particle filters can be used to reduce the computational cost to linear in the number of observations, at the expense of introducing small errors; and propose two new, optimum resampling algorithms for this problem. One, a version of rejection control, allows the particle filter to automatically choose the number of particles required at each time-step. The new resampling algorithms substantially out-perform standard resampling algorithms on examples we consider; and we demonstrate how the resulting particle filter is practicable for segmentation of human GC content.

AB - We propose an on-line algorithm for exact filtering of multiple changepoint problems. This algorithm enables simulation from the true joint posterior distribution of the number and position of the changepoints for a class of changepoint models. The computational cost of this exact algorithm is quadratic in the number of observations. We further show how resampling ideas from particle filters can be used to reduce the computational cost to linear in the number of observations, at the expense of introducing small errors; and propose two new, optimum resampling algorithms for this problem. One, a version of rejection control, allows the particle filter to automatically choose the number of particles required at each time-step. The new resampling algorithms substantially out-perform standard resampling algorithms on examples we consider; and we demonstrate how the resulting particle filter is practicable for segmentation of human GC content.

KW - Direct simulation

KW - Isochores

KW - Rejection control

KW - Sequential Monte Carlo

KW - Stratified sampling

KW - Particle Filtering

U2 - 10.1111/j.1467-9868.2007.00601.x

DO - 10.1111/j.1467-9868.2007.00601.x

M3 - Journal article

VL - 69

SP - 589

EP - 605

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 - 4

ER -