Minimax and pointwise sequential changepoint detection and identification for general stochastic models

Minimax and pointwise sequential changepoint detection and identification for general stochastic models

This paper considers the problem of joint change detection and identification assuming multiple composite post-change hypotheses. We propose a multihypothesis changepoint detection-identification procedure that controls the probabilities of false alarm and wrong identification. We show that the prop...

Full description

Bibliographic Details
Published in:Journal of multivariate analysis Vol. 190. P. 104977
Main Author: Pergamenshchikov, Serguei M.
Other Authors: Tartakovsky, Alexander G., Spivak, Valentin S.
Format: Article
Language:English
Subjects:
асимптотическая оптимальность
стохастические модели
эргодические марковские процессы
статьи в журналах
Online Access:http://vital.lib.tsu.ru/vital/access/manager/Repository/koha:001000951
LEADER 02147nab a2200313 c 4500
001 koha001000951
005 20230426130152.0
007 cr |
008 230425|2022 xxu s a eng d
024 7 |a 10.1016/j.jmva.2022.104977  |2 doi 
035 |a koha001000951 
040 |a RU-ToGU  |b rus  |c RU-ToGU 
100 1 |a Pergamenshchikov, Serguei M. 
245 1 0 |a Minimax and pointwise sequential changepoint detection and identification for general stochastic models  |c S. M. Pergamenshchikov, A. G. Tartakovsky, V. S. Spivak 
336 |a Текст 
337 |a электронный 
504 |a Библиогр.: 18 назв. 
520 3 |a This paper considers the problem of joint change detection and identification assuming multiple composite post-change hypotheses. We propose a multihypothesis changepoint detection-identification procedure that controls the probabilities of false alarm and wrong identification. We show that the proposed procedure is asymptotically minimax and pointwise optimal, minimizing moments of the detection delay as probabilities of false alarm and wrong identification approach zero. The asymptotic optimality properties hold for general stochastic models with dependent and nonidentically distributed observations. We illustrate general results for detection-identification of changes in multistream Markov ergodic processes. We consider several examples, including an application to rapid detection-identification of COVID-19 in Italy. Our proposed sequential algorithm allows much faster detection of COVID-19 than standard methods. 
653 |a асимптотическая оптимальность 
653 |a стохастические модели 
653 |a эргодические марковские процессы 
655 4 |a статьи в журналах 
700 1 |a Tartakovsky, Alexander G. 
700 1 |a Spivak, Valentin S. 
773 0 |t Journal of multivariate analysis  |d 2022  |g Vol. 190. P. 104977  |x 0047-259X 
852 4 |a RU-ToGU 
856 4 |u http://vital.lib.tsu.ru/vital/access/manager/Repository/koha:001000951 
908 |a статья 
999 |c 1000951  |d 1000951 

Similar Items