bibtype J - Journal Article
ARLID 0410646
utime 20240103182228.9
mtime 20060210235959.9
title (primary) (eng) Large deviations of divergence measures on partitions
specification
page_count 16 s.
serial
ARLID cav_un_epca*0257116
ISSN 0378-3758
title Journal of Statistical Planning and Inference
volume_id 93
page_num 1-16
publisher
name Elsevier
keyword large deviations
keyword partitions
keyword goodness-of-fit
author (primary)
ARLID cav_un_auth*0212817
name1 Beirlant
name2 J.
country BE
author
ARLID cav_un_auth*0212818
name1 Devroye
name2 L.
country CA
author
ARLID cav_un_auth*0211826
name1 Györfi
name2 L.
country HU
author
ARLID cav_un_auth*0101218
name1 Vajda
name2 Igor
institution UTIA-B
fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
COSATI 12B
cas_special
project
project_id GA102/99/1137
agency GA ČR
ARLID cav_un_auth*0004432
project
project_id 579
agency Copernicus
country XE
research AV0Z1075907
abstract (eng) We discuss Chernoff-type large deviation results for the total variation, the I-divergence errors, and the chi-square-divergence errors on partitions. In contrast to the total variation and the I-divergence, the chi-square-divergence has an unconventional large deviation rate. Applications to Bahadur efficiencies of goodness-of-fit tests based on these divergence measures for multivariate observations are given.
RIV BB
department SI
permalink http://hdl.handle.net/11104/0130734
ID_orig UTIA-B 20010115
arlyear 2001
mrcbU63 cav_un_epca*0257116 Journal of Statistical Planning and Inference 0378-3758 1873-1171 Roč. 93 1/2 2001 1 16 Elsevier