Pettitt’s change point test is only applicable for detecting a single change point in a time series. To analyze multiple change points in a time series, you can use methods such as binary segmentation, dynamic programming, and Bayesian change point analysis.
Many options are possible. If you are using R, let me borrow from the CRAN task view of time series analysis:
Change point detection is provided in strucchange and strucchangeRcpp (using linear regression models) and in trend (using nonparametric tests). The changepoint package provides many popular changepoint methods, and ecp does nonparametric changepoint detection for univariate and multivariate series. changepoint.np implements the nonparametric PELT algorithm, changepoint.mv detects changepoints in multivariate time series, while changepoint.geo implements the high-dimensional changepoint detection method GeomCP. Factor-augmented VAR (FAVAR) models are estimated by a Bayesian method with FAVAR. InspectChangepoint uses sparse projection to estimate changepoints in high-dimensional time series. Rbeast provides Bayesian change-point detection and time series decomposition.
Of these, the Rbeast package is maintained by me and also available in Matlab and Python. If interested, see more infor at https://github.com/zhaokg/Rbeast.
- Badges
- Science topic