Most of the statistical estimation procedures are based on a quite simple principle: find the distribution that, within a certain class, is as similar as possible to the empirical distribution, obtained from the sample observations. This leads to the minimization of some statistical functionals, usually interpreted ad measures of distance or divergence between distributions. In this paper we study the majorization pre-order of the distance between distributions. This concept, known in literature as relative majorization, is extended to the weak definition of majorization, which is more relevant in many practical contexts such as estimation problem. Providing mathematical proofs, we study under which conditions statistical functionals are consistent with respect to the relative weak majorization (from above) pre-order.

(2014). Divergence Measures and Weak Majorization in Estimation Problems [conference presentation - intervento a convegno]. Retrieved from http://hdl.handle.net/10446/30937

Divergence Measures and Weak Majorization in Estimation Problems

LANDO, Tommaso;BERTOLI BARSOTTI, Lucio
2014-01-01

Abstract

Most of the statistical estimation procedures are based on a quite simple principle: find the distribution that, within a certain class, is as similar as possible to the empirical distribution, obtained from the sample observations. This leads to the minimization of some statistical functionals, usually interpreted ad measures of distance or divergence between distributions. In this paper we study the majorization pre-order of the distance between distributions. This concept, known in literature as relative majorization, is extended to the weak definition of majorization, which is more relevant in many practical contexts such as estimation problem. Providing mathematical proofs, we study under which conditions statistical functionals are consistent with respect to the relative weak majorization (from above) pre-order.
2014
Lando, Tommaso; BERTOLI BARSOTTI, Lucio
File allegato/i alla scheda:
File Dimensione del file Formato  
MATH-20 paper proceedings pubblicato.pdf

Solo gestori di archivio

Descrizione: publisher's version - versione dell'editore
Versione: publisher's version - versione editoriale
Licenza: Licenza default Aisberg
Dimensione del file 412.22 kB
Formato Adobe PDF
412.22 kB Adobe PDF   Visualizza/Apri
Pubblicazioni consigliate

Aisberg ©2008 Servizi bibliotecari, Università degli studi di Bergamo | Terms of use/Condizioni di utilizzo

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/10446/30937
Citazioni
  • Scopus ND
  • ???jsp.display-item.citation.isi??? ND
social impact