Tools for outlier and structural changes detection in time series analysis using Bayesian Dynamic Linear Model.
Project description
Welcome to pybats-detection
The pybats-detection is a python package with routines implemented in python for detection of outlier and structural changes in time series using Bayesian Dynamic Linear Models (DLM).
The currently version of the package implements the automatic monitoring, manual intervention and smoothing for DLM’s.
The stable version of pybats-detection can be installed from PyPI using:
pip install pybats-detection
The development version can be installed from GitHub using:
git clone git@github.com:Murabei-OpenSource-Codes/develop/pybats-detection.git pybats-detection
cd pybats-detection
python setup.py install
The package uses the pybats.dglm.dlm objects from PyBATS package as an input for the following classes:
-
Monitoring: perform automatic monitoring of outlier and/or structural changes in time series according to West and Harisson (1986) . -
Intervention: perform manual intervention of outlier and/or structural changes in time series according to West and Harrison (1989). -
Smoothing: compute the retrospective state space parameter and predictive distributions.
All three classes have the fit method which received the univariate time series
as a pandas.Series object and further arguments related to each class.
User manuals can be found in:
-
pybats_detection: detailed explanation of
pybats-detectionusability. -
quick_start: quick reference guide with step-by-step usability.
Authors
pybats-detection was developed by André Menezes and
Eduardo Gabriel
while working as Data Scientist at Murabei Data Science
advised by professor Hélio Migon and
André Baceti .
License
The pybats-detection package is released under the Apache License, Version 2.0.
Please, see file LICENSE.md.
Download files
Download the file for your platform. If you're not sure which to choose, learn more about installing packages.
Source Distribution
Built Distribution
File details
Details for the file pybats-detection-0.2.1.tar.gz.
File metadata
- Download URL: pybats-detection-0.2.1.tar.gz
- Upload date:
- Size: 19.8 kB
- Tags: Source
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.1.1 pkginfo/1.4.2 requests/2.22.0 setuptools/45.2.0 requests-toolbelt/0.8.0 tqdm/4.30.0 CPython/3.8.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | b023e3a21a95c1bf070863018fb458aff5dcf0645d313d5844e97c40909b5042 |
|
| MD5 | edabe907ebe0f83ba05cf1514c28f0ab |
|
| BLAKE2b-256 | 4ea540b243a150fa3c53783bac3acd8fc5fb317ed538a80ce207ced73a720e92 |
File details
Details for the file pybats_detection-0.2.1-py3-none-any.whl.
File metadata
- Download URL: pybats_detection-0.2.1-py3-none-any.whl
- Upload date:
- Size: 26.0 kB
- Tags: Python 3
- Uploaded using Trusted Publishing? No
- Uploaded via: twine/3.1.1 pkginfo/1.4.2 requests/2.22.0 setuptools/45.2.0 requests-toolbelt/0.8.0 tqdm/4.30.0 CPython/3.8.10
File hashes
| Algorithm | Hash digest | |
|---|---|---|
| SHA256 | 56531c186011ac2142652b34b2fccb0f8d89663f504a38030105b7020a62abc3 |
|
| MD5 | d5059213bb23569b0c5416389da424e3 |
|
| BLAKE2b-256 | 710d0d2554dac9480e03a4817fd6883f186e828c9634502ffc5b25bd4541d620 |
