Original title: Modifications of divergence criteria in continuous families
Translated title: Modifikace divergenčních kriterií ve spojitých rodinách
Authors: Vajda, Igor
Document type: Research reports
Year: 2008
Language: eng
Series: Research Report, volume: 2230
Abstract: [eng] [cze]
The paper extends the applicability of the minimum divergence criterion to the parameter estimation in continuous statistical models. Four criteria are motivated, all of them leading to different Fisher consistent estimators: Superdivergence, Subdivergence, Disturbed power divergence, Disturbed Rényi divergence. Článek rozšiřuje aplikace dibergenčního kriteria na odhadování parametrů spojitých statistických modelů. Jsou zdůvodněny čtyři kriteria vedoucí k obecně různým Fisherovsky konzistentním odhadům: Superdivergence, Subdivergence, Porušené mocninné divergence a Porušené Rényiho divergence.
Keywords: Deficit of accuracy; Inaccuracy; Minimum divergence estimation; Optimal estimation
Project no.: CEZ:AV0Z10750506 (CEP), GA102/07/1131 (CEP), 1M0572 (CEP)
Funding provider: GA ČR, GA MŠk

Institution: Institute of Information Theory and Automation AS ČR (web)
Document availability information: Fulltext is available at external website.
External URL: http://library.utia.cas.cz/separaty/2008/SI/vajda-modifications of divergence criteria in continuous families.pdf
Original record: http://hdl.handle.net/11104/0167254

Permalink: http://www.nusl.cz/ntk/nusl-39623


The record appears in these collections:
Research > Institutes ASCR > Institute of Information Theory and Automation
Reports > Research reports