Efficient construction of reversible jump Markov chain Monte Carlo proposal distributions

By Steve P. Brooks, Paolo Giudici, Gareth O. Roberts. 2003

325 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/ca062942e42c475e7312d683a3107a4d60f985ff

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Paul Fearnhead
2. Andrew Gelman
3. David I. Hastie
4. Peter J. Green
5. Nial Friel
6. Galin L. Jones
7. Anthony N. Pettitt
8. Xiao-Li Meng

In Statistics and Probability

1. Xiao-Li Meng
2. Andrew Gelman
3. Paul Fearnhead
4. Galin L. Jones
5. Steve P. Brooks
6. David I. Hastie
7. Peter J. Green
8. Anthony N. Pettitt

Citations in the corpus

• Model choice using reversible jump Markov chain Monte Carlo (2012, background): “As in the saturated state space of Brooks et al. (2003b) (see section 4), the idea is to work on a more general state space, where the simulation keeps track of all k rather than only the current one.”
• Trans-dimensional inverse problems, model comparison and the evidence (2006, background): “Although the approach is gaining recognition there remain important practical issues to be addressed, not least of which is the design of efficient proposal distributions for the dimension jump (see Brooks et al. 2003).”
• Handbook of Markov Chain Monte Carlo: Hardcover: 619 pages Publisher: Chapman and Hall/CRC Press (first edition, May 2011) Language: English ISBN-10: 1420079417 (2012, methodology): “Brooks et al. (2003c), Ehlers and Brooks (2003), and Vermaak et al. (2004) each detail descriptions on the use of reversible jump samplers for this class of problems.”
• Seismic tomography with the reversible jump algorithm (2009, background): “Recent developments in acoustics (e.g. Derode et al. 2003) and seismology (Campillo & Paul 2003) showed that it is possible to perform the cross-correlation between signals recorded at two stations and extract Green’s function (the signal that would be recorded at one station if an impulsive force…”
• Markov chain Monte Carlo (MCMC) sampling methods to determine optimal models, model resolution and model choice for Earth Science problems (2009, background): “If the proposal function is too conservative (i.e. very small changes to a model) then the sampler moves very slowly around the model space and has an acceptance which is too high (a rule of thumb is about 25–30% of models should be accepted, Gilks et al., 1996; Brooks et al., 2003).”
• Output Assessment for Monte Carlo Simulations via the Score Statistic (2006, unknown): “See Brooks, Giudici, and Roberts (2003), for example.”
• Exact and efficient Bayesian inference for multiple changepoint problems (2006, methodology): “Potential difficulties of this approach include designing moves, particular ones between different models, which enable the MCMC algorithm to mix well (for guidelines on designing reversible jump MCMC algorithms see Brooks et al. , 2003), and being able to detect convergence of the algorithm.”
• Marginal likelihood estimation via power posteriors (2008, background): “However, recent work in Brooks et al. (2003) has tackled this problem to some extent.”