hal-00455803, version 1
On the kernel rule for function classification
Annals of the institute of mathematical statistics (2006) 619-633
Abstract: Let X be a random variable taking values in a function space F, and let Y be a discrete random label with values 0 and 1. We investigate asymptotic properties of the moving window classification rule based on independent copies of the pair (X, Y ). Contrary to the finite dimensional case, it is shown that the moving window classifier is not universally consistent in the sense that its probability of error may not converge to the Bayes risk for some distributions of (X, Y ). Sufficient conditions both on the space F and the distribution of X are then given to ensure consistency.
- 1:
- Institut national de la recherche agronomique (INRA) : UR0729
- 2:
- CNRS : UMR7599 – Université Pierre et Marie Curie [UPMC] - Paris VI – Université Paris VII - Paris Diderot
- 3:
- Université Pierre et Marie Curie [UPMC] - Paris VI
- 4:
- CNRS : UMR6625 – Université de Rennes 1 – École normale supérieure de Cachan - ENS Cachan – Institut National des Sciences Appliquées (INSA) : - RENNES – Université de Rennes II - Haute Bretagne
- Domain : Mathematics/Statistics
Statistics/Statistics Theory
- hal-00455803, version 1
- http://hal.archives-ouvertes.fr/hal-00455803
- oai:hal.archives-ouvertes.fr:hal-00455803
- From:
- Submitted on: Thursday, 11 February 2010 11:41:02
- Updated on: Thursday, 18 February 2010 17:30:14



Associated documents
Export