Minimum power divergence estimators, maximum likelihood and exponential families

Abstract : This note introduces the dual representation of the divergence between two distributions in a parametric model. Resulting estimators for the divergence as for the parameter are derived. These estimators do not make use of any grouping nor smoothing. It is proved that all power-type divergences with non negative index induce the same estimator of the parameter on any exponential family, which is the MLE.
Document type :
Preprints, Working Papers, ...
Complete list of metadatas

https://hal.sorbonne-universite.fr/hal-00613126
Contributor : Michel Broniatowski <>
Submitted on : Tuesday, August 2, 2011 - 10:07:00 PM
Last modification on : Wednesday, March 21, 2018 - 6:56:47 PM
Long-term archiving on : Thursday, November 3, 2011 - 2:21:55 AM

Files

31072011.pdf
Files produced by the author(s)

Identifiers

  • HAL Id : hal-00613126, version 1
  • ARXIV : 1108.0772

Citation

Michel Broniatowski. Minimum power divergence estimators, maximum likelihood and exponential families. 2011. ⟨hal-00613126v1⟩

Share

Metrics

Record views

20

Files downloads

18