2021/02/01 by Alexander G. Tartakovsky, Tartakovsky, Alexander G.
Decision Sciences · #62L15 #Advanced Statistical Process Monitoring #FOS: Mathematics #Probability (math.PR) #Statistics Theory (math.ST)
paper · pdf · doi:10.48550/arxiv.2102.01306
openalex publication_date 2021/02/01 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28
The paper addresses a joint sequential changepoint detection and\nidentification/isolation problem for a general stochastic model, assuming that\nthe observed data may be dependent and non-identically distributed, the prior\ndistribution of the change point is arbitrary, and the post-change hypotheses\nare composite. The developed detection-identification theory generalizes the\nchangepoint detection theory developed by Tartakovsky (2019) to the case of\nmultiple composite post-change hypotheses when one has not only to detect a\nchange as quickly as possible but also to identify (or isolate) the true\npost-change distribution. We propose a multi-hypothesis change\ndetection-identification rule and show that it is nearly optimal, minimizing\nmoments of the delay to detection as the probability of a false alarm and the\nprobabilities of misidentification go to zero.\n