Detection and Estimation of Multiple Transient Changes

TL;DR

This paper introduces new likelihood-based sequential and retrospective methods for detecting and estimating multiple transient change-points in processes, effectively controlling false alarms and readjustments.

Contribution

It presents novel statistical tools specifically designed for transient change detection, with improved accuracy and control over false rates compared to existing methods.

Findings

Methods effectively detect transient changes

Achieve simultaneous control of false alarms and readjustments

Estimate change-points with high accuracy

Abstract

Change-point detection methods are proposed for the case of temporary failures, or transient changes, when an unexpected disorder is ultimately followed by a readjustment and return to the initial state. A base distribution of the "in-control" state changes to an "out-of-control" distribution for unknown periods of time. Likelihood based sequential and retrospective tools are proposed for the detection and estimation of each pair of change-points. The accuracy of the obtained change-point estimates is assessed. Proposed methods offer simultaneous control the familywise false alarm and false readjustment rates at the pre-chosen levels.