Detalles del recurso

Descripción

In this article, we develop a modern perspective on Akaike's information criterion and Mallows's Cp for model selection, and propose generalisations to spherically and elliptically symmetric distributions. Despite the differences in their respective motivation, Cp and Akaike's information criterion are equivalent in the special case of Gaussian linear regression. In this case, they are also equivalent to a third criterion, an unbiased estimator of the quadratic prediction loss, derived from loss estimation theory. We then show that the form of the unbiased estimator of the quadratic prediction loss under a Gaussian assumption still holds under a more general distributional assumption, the family of spherically symmetric distributions. One of the features of our results is that our criterion does not rely on the specificity of the distribution, but only on its spherical symmetry. The same kind of criterion can be derived for a family of elliptically contoured distribution, which allows correlations, when considering the invariant loss. More specifically, the unbiasedness property is relative to a distribution associated to the original density.

Pertenece a

Archive EduTice a CCSD electronic archive server based on P.A.O.L  

Autor(es)

Boisbunon, Aurélie -  Canu, Stephane -  Fourdrinier, Dominique -  Strawderman, William -  Wells, Martin T. - 

Id.: 70713687

Idioma: inglés  - 

Versión: 1.0

Estado: Final

Palabras clave[MATH.MATH -  ST] Mathematics [math]/Statistics [math.ST] - 

Tipo de recurso: info:eu-repo/semantics/preprint  -  Preprints, Working Papers, ...  - 

Tipo de Interactividad: Expositivo

Nivel de Interactividad: muy bajo

Audiencia: Estudiante  -  Profesor  -  Autor  - 

Estructura: Atomic

Coste: no

Copyright: sí

Requerimientos técnicos:  Browser: Any - 

Relación: [References] info:eu-repo/semantics/altIdentifier/arxiv/1308.2766

Fecha de contribución: 09-ene-2018

Contacto:

Localización:
* hal-00851206

Otros recursos del mismo autor(es)

  1. Confidence intervals for the means of the selected populations Consider an experiment in which $p$ independent populations $\pi_{i}$ with corresponding unknown mea...
  2. Procédé d'élaboration d'un discriminateur multiclasse
  3. ESTIMATION OF MINIMUM MEASURE SETS IN REPRODUCING KERNEL HILBERT SPACES AND APPLICATIONS. International audience
  4. Recovering sparse signals with non-convex penalties and DC programming This paper considers the problem of recovering a sparse signal representation according to a signal ...
  5. Functional learning through kernels This paper reviews the functional aspects of statistical learning theory. The main point under consi...

Otros recursos de la mismacolección

  1. A Probabilistic Analysis of Some Tree Algorithms November 2004
  2. On the asymptotic behavior of some Algorithms November 2004
  3. Formation sur des réseaux professionnels Listes de diffusion pour enseignants du premier degré Les listes de diffusion pédagogiques permettent aux enseignants de mutualiser des ressources, de coo...
  4. Les figures du hasard dans l'oeuvre de Pirandello
  5. $rec.titulo Papier présenté au premier congrès de l'AFS (Association française de sociologie) dans le cadre du R...

Aviso de cookies: Usamos cookies propias y de terceros para mejorar nuestros servicios, para análisis estadístico y para mostrarle publicidad. Si continua navegando consideramos que acepta su uso en los términos establecidos en la Política de cookies.