Fast on-line changepoint detection using heavily-weighted CUSUM and veto-based decision rules
基于重加权 CUSUM 与否决式决策规则的快速在线变点检测
We study on-line changepoint detection in the context of a linear regression model, developing two novel contributions. Firstly, we propose a class of heavily weighted statistics based on the CUSUM process of the regression residuals, which are specifically designed to ensure timely detection of breaks occurring early on during the monitoring horizon. Secondly, we develop a class of composite statistics using different weighting schemes; the decision rule to mark a…