Publications-Periodical Articles

Article View/Open

Publication Export

Google ScholarTM

NCCU Library

Citation Infomation

Related Publications in TAIR

題名 Change point detection with adaptive sampling for binary responses
作者 楊素芬
Yi, Yanqing;Yang, Su-Fen
貢獻者 統計系
關鍵詞 Adaptive sampling; Change point detection; Markov decision process; Exact distribution; False alarm rate; Statistical power
日期 2026-07
上傳時間 14-Jul-2026 11:36:02 (UTC+8)
摘要 We propose using an adaptive sampling method to detect change points for a system with multiple lines. The adaptive sampling utilizes the information in responses to learn on which line is more likely to have a change thus allocating more units to the line. The learning process is formatted as a Markov decision process and the optimal sampling is approximated by using the Bellman operator iteratively based on the average reward criterion. We demonstrate the performance of the proposed method for binary responses using the exact distribution method for adaptive sampling. Our numeric results show that the adaptive sampling samples more often the line that has a change and the statistical power to detect a change is generally better than those with a bandit UCB method and the equal randomization except for a small sample size of 10 for the equal randomization. When sample sizes increase or the difference between out-of-control and in-control probabilities increases, the adaptive sampling allocates higher proportion of units averagely to the line with a change and the statistical power to detect a change increases. The statistical power under the bandit UCB is close to those with adaptive sampling when the sample size increases to 100 for a large difference between out-of control and in-control probabilities.
關聯 Statistical Papers, Vol.67, article number 92, 19 pages
資料類型 article
DOI https://doi.org/10.1007/s00362-026-01873-4
dc.contributor 統計系
dc.creator (作者) 楊素芬
dc.creator (作者) Yi, Yanqing;Yang, Su-Fen
dc.date (日期) 2026-07
dc.date.accessioned 14-Jul-2026 11:36:02 (UTC+8)-
dc.date.available 14-Jul-2026 11:36:02 (UTC+8)-
dc.date.issued (上傳時間) 14-Jul-2026 11:36:02 (UTC+8)-
dc.identifier.uri (URI) https://ah.lib.nccu.edu.tw/item?item_id=183303-
dc.description.abstract (摘要) We propose using an adaptive sampling method to detect change points for a system with multiple lines. The adaptive sampling utilizes the information in responses to learn on which line is more likely to have a change thus allocating more units to the line. The learning process is formatted as a Markov decision process and the optimal sampling is approximated by using the Bellman operator iteratively based on the average reward criterion. We demonstrate the performance of the proposed method for binary responses using the exact distribution method for adaptive sampling. Our numeric results show that the adaptive sampling samples more often the line that has a change and the statistical power to detect a change is generally better than those with a bandit UCB method and the equal randomization except for a small sample size of 10 for the equal randomization. When sample sizes increase or the difference between out-of-control and in-control probabilities increases, the adaptive sampling allocates higher proportion of units averagely to the line with a change and the statistical power to detect a change increases. The statistical power under the bandit UCB is close to those with adaptive sampling when the sample size increases to 100 for a large difference between out-of control and in-control probabilities.
dc.format.extent 106 bytes-
dc.format.mimetype text/html-
dc.relation (關聯) Statistical Papers, Vol.67, article number 92, 19 pages
dc.subject (關鍵詞) Adaptive sampling; Change point detection; Markov decision process; Exact distribution; False alarm rate; Statistical power
dc.title (題名) Change point detection with adaptive sampling for binary responses
dc.type (資料類型) article
dc.identifier.doi (DOI) 10.1007/s00362-026-01873-4
dc.doi.uri (DOI) https://doi.org/10.1007/s00362-026-01873-4