Skip to main content

Methods that attempt to detect when a change occurs in a distribution, process, or function.

Change point detection methods attempt to detect when a change occurs in a distribution, process, or function. For example, the function relating X and Y may have one slope until a given point, and a different slope thereafter, where the point is not known a-priori, but determining it is the goal of an analysis. A common situation where change point detection might arise is with time series data.