Bayesian Wavelet Change Point Detection
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Type Name Latest commit message Commit time
Failed to load latest commit information.


(Bayesian) Wavelet Change Point Detection

Detect the change point in mean in high-dimensional time series using dimension reduction JLDetectChangePoint followed by a wavelet technique for estimating the change point. For lower dimensional time series, use detectChangePoint, which does not do dimension reduction. Currently, the Bayesian aspect of this algorithm is not implemented, and the only prior allowed is a uniform prior.

Change points of time series with values in arbitrary metric spaces are also detected via metricChangePoint. Finally, change points in persistence diagrams of spatial data are detected using persistenceChangePoint. Unfortunately, the TDA library that is required for persistenceChangePoint is not working with R 3.4 for some Mac users. It does work on R 3.3.2 (Sincere Pumpkin Patch), which is why this package has only been tested relative to R 3.3.2. I delayed releasing this in hopes that TDA would get fixed.

To install, use devtools::install_github("speegled/cpbaywave").

I would love to have contributers help with this package. To contribute, please make a pull request on the github repo If you have an idea for a major change or update, consider creating an issue describing your idea first. I also have more ideas for improvements than time (as of July 3, 2017), so if you have some free time right and want to contribute, please contact me.

Code of Conduct: All contributors to this package will treat other contributors with respect.