Sequential change-point detection when unknown parameters are present in the pre-change distribution

DOI10.1214/009053605000000859zbMATH Open1091.62064arXivmath/0605322OpenAlexW3101806252MaRDI QIDQ2493547FDOQ2493547

Authors: Yajun Mei 

Publication date: 21 June 2006

Published in: The Annals of Statistics (Search for Journal in Brave)

Abstract: In the sequential change-point detection literature, most research specifies a required frequency of false alarms at a given pre-change distribution $f_{heta}$ and tries to minimize the detection delay for every possible post-change distribution $g_{lambda}$. In this paper, motivated by a number of practical examples, we first consider the reverse question by specifying a required detection delay at a given post-change distribution and trying to minimize the frequency of false alarms for every possible pre-change distribution $f_{heta}$. We present asymptotically optimal procedures for one-parameter exponential families. Next, we develop a general theory for change-point problems when both the pre-change distribution $f_{heta}$ and the post-change distribution $g_{lambda}$ involve unknown parameters. We also apply our approach to the special case of detecting shifts in the mean of independent normal observations.

Full work available at URL: https://arxiv.org/abs/math/0605322

Recommendations

• Sequential change-point detection when the pre- and post-change parameters are unknown
• Sequential change detection in the presence of unknown parameters
• Using the generalized likelihood ratio statistic for sequential detection of a change-point
• A generalized cusum procedure for sequential detection of change-point in a parametric family when the initial parameter is unknown
• Detection of changes in the mean of a sequence of random variables

zbMATH Keywords

surveillancestatistical process controlquality controlchange-pointasymptotic optimalityoptimizerpower one tests

Mathematics Subject Classification ID

Applications of statistics in engineering and industry; control charts (62P30) Asymptotic properties of parametric tests (62F05) Sequential statistical analysis (62L10) Optimal stopping in statistics (62L15)

Cites Work

• Optimal stopping times for detecting changes in distributions
• Title not available (Why is that?)
• Title not available (Why is that?)
• Sequential analysis. Tests and confidence intervals
• Sequential analysis: Some classical problems and new challenges. (With comments and rejoinder).
• Information bounds and quick detection of parameter changes in stochastic systems
• Title not available (Why is that?)
• Procedures for Reacting to a Change in Distribution
• Optimum Character of the Sequential Probability Ratio Test
• CONTINUOUS INSPECTION SCHEMES
• Decision theoretic optimality of the cusum procedure
• Optimal detection of a change in distribution
• Open-ended tests for Koopman-Darmois families
• Using the generalized likelihood ratio statistic for sequential detection of a change-point
• On Optimum Methods in Quickest Detection Problems
• Title not available (Why is that?)
• A diffusion process and its applications to detecting a change in the drift of Brownian motion
• On Excess Over the Boundary
• Extended renewal theory and moment convergence in Anscombe's theorem
• Bioequivalence trials, intersection-union tests and equivalence confidence sets. With comments and a rejoinder by the authors
• Average run lengths of an optimal method of detecting a change in distribution
• Asymptotically Optimum Sequential Inference and Design
• Optimality and almost optimality of mixture stopping rules
• On the average run length to false alarm in surveillance problems which possess an invariance structure
• Surveillance of a Simple Linear Regression
• Average run length to false alarm for surveillance schemes designed with partially specified pre-change distribution
• A robust surveillance scheme for stochastically ordered alternatives
• Sequential detection of a change in a normal mean when the initial value is unknown
• Detecting a change in regression: First-order optimality
• The Variance of One-Sided Stopping Rules

Cited In (32)

• Two tests for sequential detection of a change-point in a nonlinear model
• Title not available (Why is that?)
• Online change detection in exponential families with unknown parameters
• Sequential nonparametric tests for a change in distribution: an application to detecting radiological anomalies
• Sequential change diagnosis revisited and the Adaptive Matrix CuSum
• Asymptotically optimal robust information-based quick detection for general stochastic models with nonparametric postchange uncertainty
• A more powerful test identifying the change in mean of functional data
• Sequential change-point detection when the pre- and post-change parameters are unknown
• Sequential monitoring of a Bernoulli sequence when the pre-change parameter is unknown
• Sequential change-point detection: computation versus statistical performance
• Sequential change-point detection in continuous time when the post-change drift is unknown
• Minimax Methods for Multihypothesis Sequential Testing and Change-Point Detection Problems
• Sequential change-point detection for skew normal distribution
• Sequential Detection of Change-Points in Linear Models
• Nonanticipating estimation applied to sequential analysis and changepoint detection
• Is Average Run Length to False Alarm Always an Informative Criterion?
• Empirical likelihood test in a posteriori change-point nonlinear model
• A Bayesian Approach to Sequential Surveillance in Exponential Families
• Nonparametric Shiryaev-Roberts change-point detection procedures based on modified empirical likelihood
• A multiple hypothesis testing approach to detection changes in distribution
• Sequential Change-Point Detection in State-Space Models
• Sequential Change-Point Detection Procedures That are Nearly Optimal and Computationally Simple
• Asymptotically optimal methods of early change-point detection
• Sequential change detection in the presence of unknown parameters
• On the Bayesian sequential change-point detection
• An empirical likelihood-based CUSUM for on-line model change detection
• Change detection for uncertain autoregressive dynamic models through nonparametric estimation
• Change-point problems: bibliography and review
• A discussion on ``Detection of intrusions in information systems by sequential change-point methods by Tartakovsky, Rozovskii, Blažek, and Kim
• Mean shift testing in correlated data
• Discussion on “Is Average Run Length to False Alarm Always an Informative Criterion?” by Yajun Mei
• Discussion on “Is Average Run Length to False Alarm Always an Informative Criterion?” by Yajun Mei

This page was built for publication: Sequential change-point detection when unknown parameters are present in the pre-change distribution

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q2493547)