Online change-point detection: A traders perspective

Abstract

The aim is to present several classes of integral-transform-based test statistics for change-point analysis, examining their asymptotic behavior and small-sample performance. These methods are then adapted for real-time monitoring, enabling the rapid detection of structural changes as new data arrive. From a trader’s perspective, such tools are critical for reacting to sudden regime shifts. Practical implementation details are highlighted, and it is illustrated how these methods can support decision-making in unstable financial environments.