Limit theorems for geometric functionals of Gibbs point processes (Q376697): Difference between revisions

A point process on \(\mathbb R^d\) \((d \geq 1)\) is considered. It can be treated as a probability measure \(\mu\) on a measurable set of all locally finite subsets \(\mathcal{X} \subset \mathbb R\). A so-called geometrical functional \(\Psi(\mathcal{X}) :=\sum_{x\in \mathcal{X}}\xi(x, \mathcal{X} )\) is determined, where \(\xi\) is a positive function, defined on pairs \((x,\mathcal{X})\). This functional can be used for rarefying an a priori given point process and its transformations. On the other hand, an a priori given consistent system of distributions of the geometrical functionals, determined on all conjunctions \(\mathcal{X} \cap \Delta\), where \(\Delta \subset \mathbb R^d\) is a precompactum, determines a probability measure of some point process. In such a way, a Gibbs point process is constructed with the function \(\xi\), which depends only on distances between points in every finite set \(\mathcal{X} \cap \Delta\), and distributions of the geometrical functionals are determined by the densities of \(A \exp (-\beta\Psi (\mathcal{X} \cap \Delta))\), where \(A, \beta\) are some thermodynamic parameters. The main results of the paper relate to a limit behavior of the geometrical functionals when the intensity of a corresponding point process tends to infinity. In particular, such a point process is a Gibbs point process. In this case, some laws of great numbers and the central limit theorem are proved for every studied geometrical functional. The obtained results are applied for the analysis of models of some physical phenomena.