Publications Scientifiques

[ Article ] IMPROVED ESTIMATORS OF BREGMAN DIVERGENCE FOR MODEL SELECTION IN SMALL SAMPLES

Date de soumission: 16-02-2019
Année de Publication: 2018
Entité/Laboratoire Autres Laboratoires
Document type : Article
Discipline(s) : Mathématiques
Titre IMPROVED ESTIMATORS OF BREGMAN DIVERGENCE FOR MODEL SELECTION IN SMALL SAMPLES
Auteurs Papa Dario [1], Nkurunziza Jean de Dieu [2], Ogouyandjou Carlos [3],
Journal: Far East Journal of Theoretical Statistics
Catégorie Journal: Internationale
Impact factor:
Volume Journal: 54
DOI: 10.17654/TS054020189
Resume Recently in [1, 2], Bromideh introduced the Kullback-Leibler Divergence (KLD) test statistic in discriminating between two models. It was found that the ratio minimized Kullback-Leibler divergence (RMKLD) works better than the ratio of maximized likelihood (RML) for small sample size. The aim of this paper is to generalize the works of Ali-Akbar Bromideh by proposing a hypothesis testing based on Bregman Divergence (BD) in order to improve the process of choice of the model. We investigate the problem of model choice and propose a unified method for model selection and estimation procedure with desired theoretical properties and computational convenience. After observing n data points of unknown density f; we firstly measure the closeness between the bias reduced kernel density estimator and the first estimated candidate model. Secondly between the bias reduced kernel density estimator and the second estimated candidate model. In these two cases, BD and the bias reduced kernel estimator [3] focuses on improving the convergence rates of kernel density estimators are used. We establish the asymptotic properties of BD estimator and approximations of the power functions are deduced. The multi-step MLE process will be used to estimate the parameters of the models. We explain the applicability of the BD by a real data set and by the data generating process (DGP). The Monte Carlo simulation and then the numerical analysis will be used to interpret the result.
Mots clés A bias reduced kernel estimator, Bregman divergence, hypothesis test
Pages 189 - 224
Fichier (PDF)

Publications par entité

  • Centre de Formation et de Recherche en matière de Population (CEFORP) (15)
  • Centre Interfacultaire de Formation et de Recherche en Environnement pour le Développement Durable (CIFRED) (18)
  • Chaire Internationale en Physique Mathématique et Applications (CIPMA-Chaire UNESCO) (16)
  • Ecole Nationale d'Administration et de Magistrature (ENAM) (44)
  • Ecole Nationle d'Economie Appliquée et de Management (72)
  • Ecole Normale Supérieure (23)
  • Ecole Polytechnique d'Abomey-Calavi (EPAC) (835)
  • Faculté de droit et de Science Politique (FADESP) (45)
  • Faculté des Lettres et Sciences Humaines (FLASH) (789)
  • Faculté des Sciences Agronomiques (1376)
  • Faculté des Sciences de la Santé (FSS) (1613)
  • Faculté des Sciences Economiques et de Gestion (FASEG) (136)
  • Faculté des Sciences et Techniques (FAST) (748)
  • Faculté des Sciences Humaines et Sociales (FASHS) (16)
  • Institut de Formation et de Recherche en Informatique (IFRI) (68)
  • Institut de Mathématiques et de Sciences Physiques (IMSP) (61)
  • Institut National de l'Eau (INE) (13)
  • INSTITUT NATIONAL DE LA JEUNESSE DE L’EDUCATION PHYSIQUE ET SPORT(INJEPS) (191)
  • Institut National des Métiers d'Art d'Archéologie et de la Culture (INMAAC) (4)
  • Institut Régional de Santé Publique (IRSP) (51)

Publications par sexe

  • Femme( 908 )
  • Homme (