Asymptotic approximation of the likelihood ...
Type de document :
Pré-publication ou Document de travail
Titre :
Asymptotic approximation of the likelihood of stationary determinantal point processes
Auteur(s) :
Poinas, Arnaud [Auteur]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Lavancier, Frédéric [Auteur]
Laboratoire de Mathématiques Jean Leray [LMJL]
Centre de Recherche en Informatique, Signal et Automatique de Lille - UMR 9189 [CRIStAL]
Lavancier, Frédéric [Auteur]
Laboratoire de Mathématiques Jean Leray [LMJL]
Discipline(s) HAL :
Mathématiques [math]/Statistiques [math.ST]
Résumé en anglais : [en]
Continuous determinantal point processes (DPPs) are a class of repulsive point processes on R^d with many statistical applications. Although an explicit expression of their density is known, this expression is too complicated ...
Lire la suite >Continuous determinantal point processes (DPPs) are a class of repulsive point processes on R^d with many statistical applications. Although an explicit expression of their density is known, this expression is too complicated to be used directly for maximum likelihood estimation. In the stationary case, an approximation using Fourier series has been suggested, but it is limited to rectangular observation windows and no theoretical results support it. In this contribution, we investigate a different way to approximate the likelihood by looking at its asymptotic behaviour when the observation window grows towards R^d. This new approximation is not limited to rectangular windows, is faster to compute than the previous one, does not require any tuning parameter, and some theoretical justifications are provided. The performances of the associated estimator are assessed in a simulation study on standard parametric models on R^d and compare favourably to common alternative estimation methods for continuous DPPs.Lire moins >
Lire la suite >Continuous determinantal point processes (DPPs) are a class of repulsive point processes on R^d with many statistical applications. Although an explicit expression of their density is known, this expression is too complicated to be used directly for maximum likelihood estimation. In the stationary case, an approximation using Fourier series has been suggested, but it is limited to rectangular observation windows and no theoretical results support it. In this contribution, we investigate a different way to approximate the likelihood by looking at its asymptotic behaviour when the observation window grows towards R^d. This new approximation is not limited to rectangular windows, is faster to compute than the previous one, does not require any tuning parameter, and some theoretical justifications are provided. The performances of the associated estimator are assessed in a simulation study on standard parametric models on R^d and compare favourably to common alternative estimation methods for continuous DPPs.Lire moins >
Langue :
Anglais
Collections :
Source :
Fichiers
- https://hal.archives-ouvertes.fr/hal-03157554/document
- Accès libre
- Accéder au document
- https://hal.archives-ouvertes.fr/hal-03157554/document
- Accès libre
- Accéder au document
- https://hal.archives-ouvertes.fr/hal-03157554/document
- Accès libre
- Accéder au document
- document
- Accès libre
- Accéder au document
- MLE2.pdf
- Accès libre
- Accéder au document