dorsal/arxiv
View SchemaBayesian Blocks: Divide and Conquer, MCMC, and Cell Coalescence Approaches
| Authors | Jeffrey D. Scargle |
|---|---|
| Categories | |
| ArXiv ID | physics/0009033 |
| URL | https://arxiv.org/abs/physics/0009033 |
Abstract
Identification of local structure in intensive data -- such as time series, images, and higher dimensional processes -- is an important problem in astronomy. Since the data are typically generated by an inhomogeneous Poisson process, an appropriate model is one that partitions the data space into cells, each of which is described by a homogeneous (constant event rate) Poisson process. It is key that the sizes and locations of the cells are determined by the data, and are not predefined or even constrained to be evenly spaced. For one-dimensional time series, the method amounts to Bayesian changepoint detection. Three approaches to solving the multiple changepoint problem are sketched, based on: (1) divide and conquer with single changepoints, (2) maximum posterior for the number of changepoints, and (3) cell coalescence. The last method starts from the Voronoi tessellation of the data, and thus should easily generalize to spaces of higher dimension.
{
"annotation_id": "0bb370b6-5c35-4c65-957f-fea9c6905b7e",
"date_created": "2026-03-02T18:00:32.779000Z",
"date_modified": "2026-03-02T18:00:32.779000Z",
"file_hash": "9896f9f852751dde4efb34790c9342dbe688afe74a8d9824b4ef9c6f84a7ba25",
"private": false,
"record": {
"abstract": "Identification of local structure in intensive data -- such as time series,\nimages, and higher dimensional processes -- is an important problem in\nastronomy. Since the data are typically generated by an inhomogeneous Poisson\nprocess, an appropriate model is one that partitions the data space into cells,\neach of which is described by a homogeneous (constant event rate) Poisson\nprocess. It is key that the sizes and locations of the cells are determined by\nthe data, and are not predefined or even constrained to be evenly spaced. For\none-dimensional time series, the method amounts to Bayesian changepoint\ndetection. Three approaches to solving the multiple changepoint problem are\nsketched, based on: (1) divide and conquer with single changepoints, (2)\nmaximum posterior for the number of changepoints, and (3) cell coalescence. The\nlast method starts from the Voronoi tessellation of the data, and thus should\neasily generalize to spaces of higher dimension.",
"arxiv_id": "physics/0009033",
"authors": [
"Jeffrey D. Scargle"
],
"categories": [
"physics.data-an",
"physics.comp-ph"
],
"title": "Bayesian Blocks: Divide and Conquer, MCMC, and Cell Coalescence Approaches",
"url": "https://arxiv.org/abs/physics/0009033"
},
"schema_id": "dorsal/arxiv",
"source": {
"execution_id": "09a26aa7-f85f-4401-a1a8-014acd7a9071",
"id": "arXiv Dataset IDs",
"type": "Model",
"variant": "snapshot-2026-03-01",
"version": "0.1.0"
},
"user_id": 1000002
}