## Bayesian Inference for Non-Markovian Point Processes (2010)

Citations: | 1 - 0 self |

### BibTeX

@MISC{Guttorp10bayesianinference,

author = {Peter Guttorp and Thordis L. Thorarinsdottir},

title = {Bayesian Inference for Non-Markovian Point Processes},

year = {2010}

}

### OpenURL

### Abstract

Statistical inference for point processes originates, as pointed out by Daley and Vere-Jones (2005), in two sources: life tables, and counting phenomena. Among early sources of inferential work are Graunt, Halley and Newton in the 18th century on the life table side, and Newcomb, Abbé and Seidel in the second half of the 19th century on the counting side (for

### Citations

935 | Reversible jump Markov chain Monte Carlo computation and Bayesian model determination
- Green
- 1995
(Show Context)
Citation Context .... Contrary to the models considered in Example 2.1, we cannot calculate the marginal likelihood (4) of a dataset x under the models M1 and M2 directly. Instead, we define a reversible jump algorithm (=-=Green, 1995-=-) where we jump between the models M1 and M2. The Bayes factor can then be obtained directly from the MCMC sample by comparing the time spent in M1 and the time spent M2. The random intensity function... |

485 | On Bayesian analysis of Mixtures with an Unknown Number of Components - Richardson, Green - 1997 |

348 | Core Team (2009) R: A Language and Environment for Statistical Computing - Development |

229 |
The Statistical Analysis of Series of Events
- Cox, Lewis
- 1966
(Show Context)
Citation Context ...bservation A, are the same as those for density estimation. Conditionally upon the total number of points N = N(A) the points are distributed as the order statistics from a distribution with density (=-=Cox and Lewis, 1966-=-) ∫ f(s) = λ(s)/ λ(u)du. (1) A Diggle (1985) used this fact to develop a kernel estimator for the intensity, and Peeling et al. (2007) used a histogram type estimator in setting up a Bayesian analysis... |

107 | Nonparametric Inference for a family of Counting Processes. The Annals of Statistics - Aalen - 1978 |

93 | An introduction to the theory of point processes. Vol. I. Probability and its Applications - Daley, Vere-Jones - 2003 |

91 | Approximte Bayesian inference for latent gaussian models using integrated MULTI-LEVEL MODELLING 335 nested laplace approximations (with discussion - Rue, Martino, et al. - 2009 |

88 | Statistical Inference and Simulation for Spatial Point Processes - Møller, Waagepetersen - 2004 |

80 | Investigations into resting-state connectivity using independent component analysis - Beckmann, DeLuca, et al. - 2005 |

72 | Simulation of nonhomogeneous Poisson processes by thinning - Lewis, Shedler - 1979 |

60 |
Some statistical methods connected with series of events
- Cox
- 1955
(Show Context)
Citation Context ...sian, and in our context serves as a prior for a nonhomogeneous Poisson process intensity, while the setup mathematically (albeit not conceptually) corresponds to a doubly stochastic Poisson process (=-=Cox, 1955-=-). The doubly stochastic Poisson process if of course of interest in its own right (see section 3). The conditional likelihood for this model, given the realisation of λ(s), s ∈ A, is simply the usual... |

42 | A kernel method for smoothing point process data - Diggle - 1985 |

39 | The spectral analysis of point processes - Bartlett - 1963 |

39 | Nonparametric Bayesian Estimation of a Spatial Poisson Intensity - Heikkinen, Arjas - 1998 |

36 |
Some tests of significance treated by the theory of probability
- Jeffreys
- 1935
(Show Context)
Citation Context ...térn model described in the previous subsection. Here, we view the data as seven independent realisations of fall precipitation events at Whiteface Mountain. 6In a Bayesian framework, Bayes factors (=-=Jeffreys, 1935-=-) offer a natural way of scoring models based on the evidence provided by the data. Specifically, suppose that p(x|θ, M) is the density function of the observed point pattern x under model M given the... |

30 |
Seismicity analysis through point-process modeling: a review
- Ogata
- 1999
(Show Context)
Citation Context ...literature. The AIC has the advantage that it can be applied to any likelihood based inference method. However, it has been noted that it tends to favour more complicated models for larger data sets (=-=Ogata, 1999-=-). This is a clear disadvantage in a setting where the modelling easily becomes computationally intractable. We discuss this issue further in Example 5.1. Bayes factors (see Example 2.1) were first us... |

27 |
Statistical analysis and modelling of spatial point patterns
- Illian
- 2008
(Show Context)
Citation Context ...ance distribution function and the spherical contact distribution function can provide the modeller with evidence for regularity or clustering in the point pattern as compared to complete randomness (=-=Illian et al., 2008-=-; Baddeley, 2010). Such comparisons can produce important guidance for choosing the correct class of models, yet these model classes are very broad, rendering the information less valuable. Statistica... |

27 | Log Gaussian cox processes - Møller, Syversveen, et al. - 1998 |

24 | Bayesian computation for nonhomogeneous Poisson processes in software reliability - Kuo, Yang - 1996 |

22 |
An introduction to the theory of point processes, 2nd edition, vol
- Daley, Vere-Jones
- 2002
(Show Context)
Citation Context ...ns used for the underlying random field are usually improper which means that the Bayes factor is only determined up to an unknown ratio of constants. 4 Cluster processes The general cluster process (=-=Daley and Vere-Jones, 2005-=-) consists of a parent point process X to each point τi of which is associated a secondary point process Zi + τi. The structure of the primary and secondary processes is a suitable source of classific... |

22 | Nonparametric Bayesian estimators for counting processes - Kim - 1999 |

22 | An estimation function approach to inference for inhomogeneous Neyman–Scott processes - Waagepetersen - 2007 |

19 | A stochastic description of precipitation - LeCam - 1961 |

18 | Markov properties of cluster processes - Baddeley, Lieshout, et al. - 1996 |

17 | Asymptotic properties of estimators for the parameters of spatial inhomogeneous poisson point processes - Cressie, Rathbun - 1994 |

17 | Handbook of Spatial Statistics - Gelfand, Diggle, et al. - 2010 |

15 |
Bayesian nonparametric statistical inference for Poisson point processes
- LO
- 1982
(Show Context)
Citation Context ...f the doubly stochastic Poisson process obtained using a log Gaussian rate function. Gutiérrez-Peña and Nieto-Barajas (2003) considered a doubly stochastic Poisson process with a gamma process (as in =-=Lo, 1982-=-) being its rate function Λ(t). This process has parameter functions α (the rate function measure) and β (the scale process). In the case of constant scale β = b, the resulting process is what they ca... |

14 | Bayesian mixture modeling for spatial Poisson process intensities, with applications to extreme value analysis - Kottas, Sanso - 2006 |

12 | Delleur, A stochastic cluster model of daily rainfall sequences - Kavvas, W - 1981 |

12 | On a class of Bayesian nonparametric estimates: Ii. hazard rate estimates - Lo, Weng - 1989 |

11 | Tractable nonparametric Bayesian inference in Poisson processes with Gaussian process intensities - Adams, Murray, et al. - 2009 |

8 | Bayes factor for non-homogeneous Poisson processes with vague prior information - Akman, Raftery - 1986 |

7 | A stochastic geometry model for functional magnetic resonance images - Hartvig - 2002 |

7 | Likelihood-based inference for Matérn Type-III repulsive point processes - Huber, Wolpert - 2009 |

7 | Dirichlet process mixtures of beta distributions, with applications to density and intensity estimation - Kottas - 2006 |

7 | Waagepetersen (2004): Statistical Inference and Simulation for Spatial Point Processes - Møller, P |

7 | Cluster priors in the Bayesian modelling of fMRI data - TASKINEN |

7 | Convergence of posteriors for discretized log Gaussian Cox processes - Waagepetersen - 2004 |

6 | A dynamic approach to the statistical analysis of point processes - Gamerman - 1992 |

6 | Likelihood-based inference for clustered line transect data - Waagepetersen, Schweder - 2006 |

5 | Rainbands, precipitation cores, and generating cells in a cyclonic storm - Hobbs, Locatelli - 1978 |

5 | Poisson point process modeling for polyphonic music transcription - Peeling, Li, et al. - 2007 |

4 | A case study on point process modelling in disease mapping. Image Analysis and Sterology - Beneš, Bodlák, et al. - 2005 |

4 | Perfect sampling for point process cluster modelling - McKeague, Loizeaux - 2002 |

4 | Bayesian nonparametric inference for mixed Poisson processes - Gutierrez-Pena, Nieto-Barajas - 2003 |

3 | Extrapolating and interpolating spatial patterns - Lieshout, J - 2002 |

3 | Spatial variation, volume 49 of Meddelanden från Statens Skogsforskningsinstitut. Statens Skogsforskningsinstitut - Matérn - 1960 |

3 | Modern Statistics for Spatial Point Processes, Foundation of the candinavian Journal of Statistics 2007. Published by Blackwell Publishing - Mφller, Waagepetersen |

2 |
Modeling Strategies
- Baddeley
- 2010
(Show Context)
Citation Context ...ction and the spherical contact distribution function can provide the modeller with evidence for regularity or clustering in the point pattern as compared to complete randomness (Illian et al., 2008; =-=Baddeley, 2010-=-). Such comparisons can produce important guidance for choosing the correct class of models, yet these model classes are very broad, rendering the information less valuable. Statistical inference for ... |

2 | A natural conjugate prior for the nonhomogeneous Poisson process with an exponential intensity function - Huang, Bier - 1999 |