Library Review

The non-homogeneous Poisson process is widely used to model the quantities software malfunction as done [Achcar et al, 1996] based at the point process (a random collection of dots in some space) in counting damage. Given M (t) is the amount of software damage observed over a period of time (1: t] and [Achcar et al, 1996] modeled M (t) using a non-homogeneous Poisson process with a function of the mean value m (t). [Morales et al., 2016] in their study entitled A Nonhomogeneous Geostatistical Poisson  Model states that the geostatistical model is for calculating data with a spacetime approach using a non-homogeneous Poisson process, the random intensity process has an additional formula with 2 (two) components, namely the Gaussian spatial component and the component for the calculation of temporal effects. In his research, Gibbs Sampling Algorithm which is an approach Markov Chain Monte Carlo (MCMC) method is used to generate sample posterior joint distribution of the parameter model. In addition, if density conditional posterior distribution is not easily identified, an algorithm can be used Metropolis-Hastings.