Auswahl der wissenschaftlichen Literatur zum Thema „Papangelou intensity“

Most finite spatial point process models specified by a density are locally stable, implying that the Papangelou intensity is bounded by some integrable function β defined on the space for the points of the process. It is possible to superpose a locally stable spatial point process X with a complementary spatial point process Y to obtain a Poisson process X ⋃ Y with intensity function β. Underlying this is a bivariate spatial birth-death process (Xt, Yt) which converges towards the distribution of (X, Y). We study the joint distribution of X and Y, and their marginal and conditional distributions. In particular, we introduce a fast and easy simulation procedure for Y conditional on X. This may be used for model checking: given a model for the Papangelou intensity of the original spatial point process, this model is used to generate the complementary process, and the resulting superposition is a Poisson process with intensity function β if and only if the true Papangelou intensity is used. Whether the superposition is actually such a Poisson process can easily be examined using well-known results and fast simulation procedures for Poisson processes. We illustrate this approach to model checking in the case of a Strauss process.

Most finite spatial point process models specified by a density are locally stable, implying that the Papangelou intensity is bounded by some integrable function β defined on the space for the points of the process. It is possible to superpose a locally stable spatial point process X with a complementary spatial point process Y to obtain a Poisson process X ⋃ Y with intensity function β. Underlying this is a bivariate spatial birth-death process (X t , Y t ) which converges towards the distribution of (X, Y). We study the joint distribution of X and Y, and their marginal and conditional distributions. In particular, we introduce a fast and easy simulation procedure for Y conditional on X. This may be used for model checking: given a model for the Papangelou intensity of the original spatial point process, this model is used to generate the complementary process, and the resulting superposition is a Poisson process with intensity function β if and only if the true Papangelou intensity is used. Whether the superposition is actually such a Poisson process can easily be examined using well-known results and fast simulation procedures for Poisson processes. We illustrate this approach to model checking in the case of a Strauss process.

AbstractWe derive moment identities for the stochastic integrals of multiparameter processes in a random-connection model based on a point process admitting a Papangelou intensity. The identities are written using sums over partitions, and they reduce to sums over non-flat partition diagrams if the multiparameter processes vanish on diagonals. As an application, we obtain general identities for the moments of k-hop counts in the random-connection model, which simplify the derivations available in the literature.

In this paper we describe methods for randomly thinning certain classes of spatial point processes. In the case of a Markov point process, the proposed method involves a dependent thinning of a spatial birth-and-death process, where clans of ancestors associated with the original points are identified, and where we simulate backwards and forwards in order to obtain the thinned process. In the case of a Cox process, a simple independent thinning technique is proposed. In both cases, the thinning results in a Poisson process if and only if the true Papangelou conditional intensity is used, and, thus, can be used as a graphical exploratory tool for inspecting the goodness-of-fit of a spatial point process model. Several examples, including clustered and inhibitive point processes, are considered.

In this paper we describe methods for randomly thinning certain classes of spatial point processes. In the case of a Markov point process, the proposed method involves a dependent thinning of a spatial birth-and-death process, where clans of ancestors associated with the original points are identified, and where we simulate backwards and forwards in order to obtain the thinned process. In the case of a Cox process, a simple independent thinning technique is proposed. In both cases, the thinning results in a Poisson process if and only if the true Papangelou conditional intensity is used, and, thus, can be used as a graphical exploratory tool for inspecting the goodness-of-fit of a spatial point process model. Several examples, including clustered and inhibitive point processes, are considered.

A new class of models for inhomogeneous spatial point processes is introduced. These locally scaled point processes are modifications of homogeneous template point processes, having the property that regions with different intensities differ only by a scale factor. This is achieved by replacing volume measures used in the density with locally scaled analogues defined by a location-dependent scaling function. The new approach is particularly appealing for modelling inhomogeneous Markov point processes. Distance-interaction and shot noise weighted Markov point processes are discussed in detail. It is shown that the locally scaled versions are again Markov and that locally the Papangelou conditional intensity of the new process behaves like that of a global scaling of the homogeneous process. Approximations are suggested that simplify calculation of the density, for example, in simulation. For sequential point processes, an alternative and simpler definition of local scaling is proposed.

A new class of models for inhomogeneous spatial point processes is introduced. These locally scaled point processes are modifications of homogeneous template point processes, having the property that regions with different intensities differ only by a scale factor. This is achieved by replacing volume measures used in the density with locally scaled analogues defined by a location-dependent scaling function. The new approach is particularly appealing for modelling inhomogeneous Markov point processes. Distance-interaction and shot noise weighted Markov point processes are discussed in detail. It is shown that the locally scaled versions are again Markov and that locally the Papangelou conditional intensity of the new process behaves like that of a global scaling of the homogeneous process. Approximations are suggested that simplify calculation of the density, for example, in simulation. For sequential point processes, an alternative and simpler definition of local scaling is proposed.

Abstract Motivated by cross-validation’s general ability to reduce overfitting and mean square error, we develop a cross-validation-based statistical theory for general point processes. It is based on the combination of two novel concepts for general point processes: cross-validation and prediction errors. Our cross-validation approach uses thinning to split a point process/pattern into pairs of training and validation sets, while our prediction errors measure discrepancy between two point processes. The new statistical approach, which may be used to model different distributional characteristics, exploits the prediction errors to measure how well a given model predicts validation sets using associated training sets. Having indicated that our new framework generalizes many existing statistical approaches, we then establish different theoretical properties for it, including large sample properties. We further recognize that non-parametric intensity estimation is an instance of Papangelou conditional intensity estimation, which we exploit to apply our new statistical theory to kernel intensity estimation. Using independent thinning-based cross-validation, we numerically show that the new approach substantially outperforms the state of the art in bandwidth selection. Finally, we carry out intensity estimation for a dataset in forestry (Euclidean domain) and a dataset in neurology (linear network).

Abstract This article derives quantitative limit theorems for multivariate Poisson and Poisson process approximations. Employing the solution of the Stein equation for Poisson random variables, we obtain an explicit bound for the multivariate Poisson approximation of random vectors in the Wasserstein distance. The bound is then utilized in the context of point processes to provide a Poisson process approximation result in terms of a new metric called $d_\pi$ , stronger than the total variation distance, defined as the supremum over all Wasserstein distances between random vectors obtained by evaluating the point processes on arbitrary collections of disjoint sets. As applications, the multivariate Poisson approximation of the sum of m-dependent Bernoulli random vectors, the Poisson process approximation of point processes of U-statistic structure, and the Poisson process approximation of point processes with Papangelou intensity are considered. Our bounds in $d_\pi$ are as good as those already available in the literature.

Les processus déterminantaux ont généré de l’intérêt dans des domaines très divers, tels que les matrices aléatoires, la théorie des processus ponctuels, ou les réseaux. Dans ce manuscrit, nous les considérons comme un type de processus ponctuel, c’est-à-dire comme un groupement de points aléatoires dans un espace très général. Ainsi, nous avons accès à une grande variété d’outils provenant de la théorie des processus ponctuels, ce qui permet une analyse précise d’un grand nombre de leur propriétés. Nous commençons par une analyse des processus déterminantaux d’un point de vue applicatif. Nous proposons ainsi différentes méthodes pour leur simulation dans un cadre général. Nous présentons une série de modèles dérivés du processus de Ginibre, et qui se trouvent être très utiles dans les applications. Troisièmement, nous introduisons un gradient différentiel sur l’espace des processus ponctuels. Grâce à des outils puissants de la théorie générale des formes de Dirichlet, nous montrons une formule d’intégration par parties pour un processus ponctuel général, et prouvons l’existence de diffusions correctement associées à ces processus. Nous sommes en mesure d’appliquer ces résultats aux processus déterminantaux, ce qui mènera à une caractérisation de ces diffusions en termes d’équations différentielles stochastiques. Enfin, nous nous intéressons au gradient différence. Dans un certain sens, nous définissons alors une intégrale de Skohorod qui satisfait une formule d'intégration par parties, c’est-à-dire que son adjoint est le gradient différence. Une application à l’étude d’une transformation aléatoire du processus ponctuel est présentée, dans laquelle nous caractérisons la distribution du processus ponctuel transformé sous certaines conditions
Determinantal point processes have sparked interest in very diverse fields, such as random matrix theory, point process theory, and networking. In this manuscript, we consider them as random point processes, i.e. a stochastic collection of points in a general space. Hence, we are granted access to a wide variety of tools from point process theory, which allows for a precise study of many of their probabilistic properties. We begin with the study of determinantal point processes from an applicative point of view. To that end, we propose different methods for their simulation in a very general setting. Moreover, we bring to light a series of models derived from the well-known Ginibre point process, which are quite suited for applications. Thirdly, we introduce a differentiable gradient on the considered probability space. Thanks to some powerful tools from Dirichlet form theory, we discuss integration by parts for general point processes, and show the existence of the associated diffusion processes correctly associated to the point processes. We are able to apply these results to the specific case of determinantal point processes, which leads us to characterizing these diffusions in terms of stochastic differential equations. Lastly, we turn our attention to the difference gradient on the same space. In a certain sense, we define a Skohorod integral, which satisfies an integration by parts formula, i.e. its adjoint is the difference operator. An application to the study of a random transformation of the point process is given, wherein we characterize the distribution of the transformed point process under mild hypotheses

Les processus déterminantaux ont généré de l’intérêt dans des domaines très divers, tels que les matrices aléatoires, la théorie des processus ponctuels, ou les réseaux. Dans ce manuscrit, nous les considérons comme un type de processus ponctuel, c’est-à-dire comme un groupement de points aléatoires dans un espace très général. Ainsi, nous avons accès à une grande variété d’outils provenant de la théorie des processus ponctuels, ce qui permet une analyse précise d’un grand nombre de leur propriétés. Nous commençons par une analyse des processus déterminantaux d’un point de vue applicatif. Nous proposons ainsi différentes méthodes pour leur simulation dans un cadre général. Nous présentons une série de modèles dérivés du processus de Ginibre, et qui se trouvent être très utiles dans les applications. Troisièmement, nous introduisons un gradient différentiel sur l’espace des processus ponctuels. Grâce à des outils puissants de la théorie générale des formes de Dirichlet, nous montrons une formule d’intégration par parties pour un processus ponctuel général, et prouvons l’existence de diffusions correctement associées à ces processus. Nous sommes en mesure d’appliquer ces résultats aux processus déterminantaux, ce qui mènera à une caractérisation de ces diffusions en termes d’équations différentielles stochastiques. Enfin, nous nous intéressons au gradient différence. Dans un certain sens, nous définissons alors une intégrale de Skohorod qui satisfait une formule d'intégration par parties, c’est-à-dire que son adjoint est le gradient différence. Une application à l’étude d’une transformation aléatoire du processus ponctuel est présentée, dans laquelle nous caractérisons la distribution du processus ponctuel transformé sous certaines conditions
Determinantal point processes have sparked interest in very diverse fields, such as random matrix theory, point process theory, and networking. In this manuscript, we consider them as random point processes, i.e. a stochastic collection of points in a general space. Hence, we are granted access to a wide variety of tools from point process theory, which allows for a precise study of many of their probabilistic properties. We begin with the study of determinantal point processes from an applicative point of view. To that end, we propose different methods for their simulation in a very general setting. Moreover, we bring to light a series of models derived from the well-known Ginibre point process, which are quite suited for applications. Thirdly, we introduce a differentiable gradient on the considered probability space. Thanks to some powerful tools from Dirichlet form theory, we discuss integration by parts for general point processes, and show the existence of the associated diffusion processes correctly associated to the point processes. We are able to apply these results to the specific case of determinantal point processes, which leads us to characterizing these diffusions in terms of stochastic differential equations. Lastly, we turn our attention to the difference gradient on the same space. In a certain sense, we define a Skohorod integral, which satisfies an integration by parts formula, i.e. its adjoint is the difference operator. An application to the study of a random transformation of the point process is given, wherein we characterize the distribution of the transformed point process under mild hypotheses

La méthode de Stein constitue une des principales techniques pour la résolution de certains problèmes d’approximation en théorie des probabilités. Dans ce manuscrit, nous l’appliquons au contexte des processus ponctuels. La première partie de ces investigations se concentre sur le processus ponctuel de Poisson. Sa propriété caractéristique d’indépendance fournit le moyen d’expliquer intuitivement pourquoi une suite de processus ponctuels de moins en moins répulsive peut converger vers un tel processus ponctuel. Ceci nous amène plus généralement à démontrer des résultats de convergence pour des suites de processus ponctuels construites à partir d’opérations telles que la superposition, l’amincissement ou l’homothétie. L’utilisation d’une distance sur les processus ponctuels, appelée distance de Kantorovich-Rubinstein, permet en outre l’obtention de taux de convergence. La seconde partie est centrée sur une classe de processus ponctuels avec beaucoup d’attractivité, appelés processus ponctuels α-stables. Leur structure basée sur un processus ponctuel de Poisson nous permet d’élargir à ces processus la méthode utilisée précédemment et de proposer de nouveaux résultats, via certaines propriétés que nous établissons sur ces processus ponctuels
Stein’s method constitutes one of the main techniques to solve some approximation problems in probability theory. In this manuscript, we apply it in the context of point processes. The first part of these investigations focuses on the Poisson point process. Its characteristic independence property provides a way to explain intuitively why a sequence of point processes becoming less and less repulsive can converge to such a point process. More generally, this leads to show some convergence results for some sequences of point processes built by several operations such as superposition, thinning and rescaling. The use of a distance on point processes, the so-called Kantorovich-Rubinstein distance, enables moreover the getting of some convergence rates. The second part is centered on a class of point processes with important attractiveness, called discrete α-stable point processes. Their structure based on a Poisson point process gives us a way to enlarge to these point processes the method used previously and to propose new results, via some properties that we state on these point processes

In this thesis, we deal with finite Gibbs point processes, especially the processes with densities with respect to a Poisson point process. The main aim of this work is to investigate a four-parametric marked point process of circular discs in three dimensions with two and three way point interactions. In the second chapter, our goal is to simulate such a process. For that purpose, the birth- death Metropolis-Hastings algorithm is presented including theoretical results. After that, the algorithm is applied on the disc process and numerical results for different choices of parameters are presented. The third chapter consists of two approaches for the estimation of parameters. First is the Takacs-Fiksel estimation procedure with a choice of weight functions as the derivatives of pseudolikelihood. The second one is the estimation procedure aiming for the optimal choice of weight functions for the estimation in order to provide better quality estimates. The theoretical background for both of these approaches is derived as well as detailed calculations for the disc process. The numerical results for both methods are presented as well as their comparison. 1