scholarly journals Erlang mixture modeling for Poisson process intensities

2021 ◽  
Vol 32 (1) ◽  
Author(s):  
Hyotae Kim ◽  
Athanasios Kottas

AbstractWe develop a prior probability model for temporal Poisson process intensities through structured mixtures of Erlang densities with common scale parameter, mixing on the integer shape parameters. The mixture weights are constructed through increments of a cumulative intensity function which is modeled nonparametrically with a gamma process prior. Such model specification provides a novel extension of Erlang mixtures for density estimation to the intensity estimation setting. The prior model structure supports general shapes for the point process intensity function, and it also enables effective handling of the Poisson process likelihood normalizing term resulting in efficient posterior simulation. The Erlang mixture modeling approach is further elaborated to develop an inference method for spatial Poisson processes. The methodology is examined relative to existing Bayesian nonparametric modeling approaches, including empirical comparison with Gaussian process prior based models, and is illustrated with synthetic and real data examples.

1995 ◽  
Vol 32 (03) ◽  
pp. 707-726 ◽  
Author(s):  
Patrick Homble ◽  
William P. McCormick

Shot noise processes form an important class of stochastic processes modeling phenomena which occur as shocks to a system and with effects that diminish over time. In this paper we present extreme value results for two cases — a homogeneous Poisson process of shocks and a non-homogeneous Poisson process with periodic intensity function. Shocks occur with a random amplitude having either a gamma or Weibull density and dissipate via a compactly supported impulse response function. This work continues work of Hsing and Teugels (1989) and Doney and O'Brien (1991) to the case of random amplitudes.


Author(s):  
CARLOS A. MOLINARES ◽  
CHRIS P. TSOKOS

The intensity function is the key entity to the power law process, also known as the Weibull process or nonhomogeneous Poisson process. It gives the rate of change of the reliability of a system as a function of time. We illustrate that a Bayesian analysis is applicable to the power law process through the intensity function. First, we show using real data, that one of the two parameters in the intensity function behaves as a random variable. With a sequence of estimates of the subject parameter we proceeded to identify the probability distribution that characterizes its behavior. Using the commonly used squared-error loss function we obtain a Bayesian reliability estimate of the power law process. Also a simulation procedure shows the superiority of the Bayesian estimate with respect to the maximum likelihood estimate and the better performance of the proposed estimate with respect to its maximum likelihood counterpart. As well, it was found that the Bayesian estimate is sensitive to a prior selection.


2001 ◽  
Vol 38 (A) ◽  
pp. 122-130 ◽  
Author(s):  
Ali S. Dabye ◽  
Yury A. Kutoyants

Consider an inhomogeneous Poisson process X on [0, T] whose unknown intensity function ‘switches' from a lower function g∗ to an upper function h∗ at some unknown point θ∗. What is known are continuous bounding functions g and h such that g∗(t) ≤ g(t) ≤ h(t) ≤ h∗(t) for 0 ≤ t ≤ T. It is shown that on the basis of n observations of the process X the maximum likelihood estimate of θ∗ is consistent for n →∞, and also that converges in law and in pth moment to limits described in terms of the unknown functions g∗ and h∗.


Sign in / Sign up

Export Citation Format

Share Document