ASCL.net

Astrophysics Source Code Library

Making codes discoverable since 1999

ASCL Code Record

[ascl:1209.001] Bayesian Blocks: Detecting and characterizing local variability in time series

Bayesian Blocks is a time-domain algorithm for detecting localized structures (bursts), revealing pulse shapes within bursts, and generally characterizing intensity variations. The input is raw time series data, in almost any form. Three data modes are elaborated: (1) time-tagged events, (2) binned counts, and (3) measurements at arbitrary times with normal errors. The output is the most probable segmentation of the observation interval into sub-intervals during which the signal is perceptibly constant, i.e. has no statistically significant variations. The idea is not that the source is deemed to actually have this discontinuous, piecewise constant form, rather that such an approximate and generic model is often useful. Treatment of data gaps, variable exposure, extension to piecewise linear and piecewise exponential representations, multi-variate time series data, analysis of variance, data on the circle, other data modes, and dispersed data are included.

This implementation is exact and replaces the greedy, approximate, and outdated algorithm implemented in BLOCK.

Code site:
https://arxiv.org/src/1207.5578v3/anc
Appears in:
http://adsabs.harvard.edu/abs/2013ApJ...764..167S
Bibcode:
2012ascl.soft09001S

Views: 3059

ascl:1209.001
Add this shield to your page
Copy the above HTML to add this shield to your code's website.