On mini-max optimality of CUSUM statistics in change point detection problem for Brownian motion
Teoriâ veroâtnostej i ee primeneniâ, Tome 61 (2016) no. 4, pp. 837-844
Voir la notice de l'article provenant de la source Math-Net.Ru
The CUSUM (CUmulative SUM) statistic is a natural generalization of the likelihood ratio. It was observed long ago that this statistic has many remarkable properties, which are useful in empirical analysis of statistical data. In this paper, we consider Lorden's minimax criterion in problems of the quickest detection of disorder, which represents the value of the drift of Brownian motion changes at an unknown and unobservable moment of time. We provide the proof of the optimality for this minimax criterion.
Keywords:
disorder, minimax criterion, two-sided inequalities for minimax risk, probabilistic characteristics
Mots-clés : Itô formula.
Mots-clés : Itô formula.
@article{TVP_2016_61_4_a10,
author = {A. N. Shiryaev},
title = {On mini-max optimality of {CUSUM} statistics in change point detection problem for {Brownian} motion},
journal = {Teori\^a vero\^atnostej i ee primeneni\^a},
pages = {837--844},
publisher = {mathdoc},
volume = {61},
number = {4},
year = {2016},
language = {ru},
url = {http://geodesic.mathdoc.fr/item/TVP_2016_61_4_a10/}
}
TY - JOUR AU - A. N. Shiryaev TI - On mini-max optimality of CUSUM statistics in change point detection problem for Brownian motion JO - Teoriâ veroâtnostej i ee primeneniâ PY - 2016 SP - 837 EP - 844 VL - 61 IS - 4 PB - mathdoc UR - http://geodesic.mathdoc.fr/item/TVP_2016_61_4_a10/ LA - ru ID - TVP_2016_61_4_a10 ER -
A. N. Shiryaev. On mini-max optimality of CUSUM statistics in change point detection problem for Brownian motion. Teoriâ veroâtnostej i ee primeneniâ, Tome 61 (2016) no. 4, pp. 837-844. http://geodesic.mathdoc.fr/item/TVP_2016_61_4_a10/