## Multiple-order Markov Chain Monte Carlo Sampling Methods with Application to a Changepoint Model (1997)

Citations: | 5 - 0 self |

### BibTeX

@TECHREPORT{Stark97multiple-ordermarkov,

author = {J. Alex Stark and William J. Fitzgerald and Stephen B. Hladky},

title = {Multiple-order Markov Chain Monte Carlo Sampling Methods with Application to a Changepoint Model},

institution = {},

year = {1997}

}

### OpenURL

### Abstract

This article explores a number of approaches to the task of formulating moves. A structure is proposed in which moves between subspaces are combined with moves within them. This is especially useful if the latter are of a Gibbs type. Ways of enhancing computational efficiency are considered. A framework is set out for reusing the results of computations in secondary moves once a primary move has been accepted or rejected. These methods are discussed in the context of a model of a piecewise-constant signal which is filtered before the addition of noise. This model was designed with a view to analysing ion channel records. Keywords: Markov chain Monte Carlo, reversible jump, changepoint analysis, ion channels

### Citations

951 | Reversible jump Markov chain Monte Carlo computation and Bayesian model determination, Biometrika 82
- Green
- 1995
(Show Context)
Citation Context ...1223 3 34967 email: jas@eng.cam.ac.uk http://www2.eng.cam.ac.uk/~jas 1 Introduction Markov chain Monte Carlo (MCMC) methods have made possible the analysis of a wide range of data models. Recent work =-=[6]-=- has developed these methods so that they can be applied to models of variable dimension. Sets of related models of differing size often have an order associated with them; this article will consider ... |

497 | On Bayesian analysis of mixtures with an unknown number of components (with discussion
- Richardson, Green
- 1997
(Show Context)
Citation Context ..., that is those where the signal is replaced by the model fit, is shown in Fig. 6(c). In this case the components of the distributions are quite distinct. If there were more overlap mixture modelling =-=[9]-=- would be an appropriate technique for extracting the conductance levels. The histogram is slightly sharper (d) when p CP is reduced by a factor of about 10 to 0:003396. When the sampler does not comp... |

81 | Hierarchical Bayesian analysis of changepoint problems - CARLIN, GELFAND, et al. - 1992 |

32 | Bayesian estimation of hidden Markov chains: a stochastic implementation - Robert, Celeux, et al. - 1993 |

19 |
Stochastic models for ion channels: introduction and bibliography
- BALL, RICE
- 1992
(Show Context)
Citation Context ...ing piecewise constant with instantaneous transitions. However, the signal has to be analogue lowpass filtered to avoid aliasing. A large amount of research has been conducted into this analysis (see =-=[2]-=- for a review); ion channel experiments are of great importance in the understanding of cell function and of signalling and control mechanisms. The methods that have been applied vary from the very si... |

19 | Estimation and Comparison of Multiple Change Point Models - Chib - 1998 |

9 |
A Bayesian restoration of an ion channel signal
- Hodgson
- 1999
(Show Context)
Citation Context ...nt of data. The underlying signal, z u (continuous line), was filtered to generate z f (dashed line), and then noise was added to generate the observed data d (isolated dots). signals by other groups =-=[1, 7]-=-. Our task was to develop an automatic model fitting scheme which was an improvement on simple methods, but which minimized the structure imposed in the analysis. This ruled out hidden Markov models a... |

3 | MCMC methods for discrete sojourn time ion channel data
- Ball, Cai, et al.
- 1996
(Show Context)
Citation Context ...nt of data. The underlying signal, z u (continuous line), was filtered to generate z f (dashed line), and then noise was added to generate the observed data d (isolated dots). signals by other groups =-=[1, 7]-=-. Our task was to develop an automatic model fitting scheme which was an improvement on simple methods, but which minimized the structure imposed in the analysis. This ruled out hidden Markov models a... |

2 | An analytical approach to changepoint detection in Laplacian noise - Wu, Fitzgerald - 1995 |

1 |
Bayesian methods applied to ion channel analysis
- Fitzgerald, Stark
- 1997
(Show Context)
Citation Context ...ince the computation involved in evaluating the primary move can be reused. 1.1 Background to the model The techniques presented in this article were developed for the analysis of ion channel records =-=[11, 5]-=-. These are the current (conductance) signals measured across the membrane of a cell using an extremely small probe. In broad terms the underlying signals are appropriately modelled as being piecewise... |

1 | The identification of discontinuities in time series using the Bayesian piecewise linear model - Ruanaidh, K, et al. - 1993 |

1 |
Bayesian modelling of patchclamp records
- Stark, Fitzgerald, et al.
- 1997
(Show Context)
Citation Context ...ince the computation involved in evaluating the primary move can be reused. 1.1 Background to the model The techniques presented in this article were developed for the analysis of ion channel records =-=[11, 5]-=-. These are the current (conductance) signals measured across the membrane of a cell using an extremely small probe. In broad terms the underlying signals are appropriately modelled as being piecewise... |