Time series co-fluctuation analysis toolbox.

A python Toolbox for time series exploratory data analysis.

What is it for?

The toolbox is designed for exploratory data analysis of Ecological Momentaty Assesment (EMA), Experience Sampling Methods (ESM), and Digital phenotyping data. The toolbox enables the inpection of time series interaction and dynamics by providing methods for summary statistics, trend and periodicity evaluation, and linear and nonlinear dependency assesment. The emphasis of the analysis is on the visualizations.

The toolbox has simple user interface, having a single configuration file for filenames, paths, and variables used in the analysis.

For the toolbox testing purposes, anomymized real life test data set is also provided.

Main Features

The toolbox features include:

• Summary statistics
• Rolling windows statistics
• Time series decomposition
• Similarity analysis
  • Similarity plot
  • Novelty score
  • Stability index
• Clustering

Installation

Pip install

pip install tscfat

You may also clone the project.

The source code is currently hosted on GitHub at: https://github.com/ArgonSilicon/tscfat/

Dependencies

The project dependencies:

• Pandas
• Numpy
• Matplotlib
• Statsmodels
• Sklearn
• Tslearn
• Nolds
• Pytest
• Seaborn

Usage example

A few motivating and useful examples of how your product can be used. Spice this up with code blocks and potentially more screenshots.

After cloning, make sure that pipenv is installed:

pip install pipenv

Activate the virtual environment:

pipenv install 

Run the example file:

pipenv run python ./Examples/example_one_subject.py

Each analysis function can be used independently. Functions assume that the input data is expected to be in a CSV file, using the following format:

• The Time column contains timestamps in unix format.
• Rest of the columns contain observations, which should be in numerical format. Each column represents one variable, rows correspond to the sampling timepoint.

For more examples and usage, please refer to the Docs.

Release History

• 0.0.1
  • Initial version, WIP

Meta

Arsi Ikäheimonen – arsi.ikaheimonen@gmail.com

Distributed under the MIT license. See LICENSE for more information.

https://github.com/ArgonSilicon/tscfat

Contributing

1. Fork it (https://github.com/ArgonSilicon/tscfat/fork)
2. Create your feature branch (git checkout -b feature/fooBar)
3. Commit your changes (git commit -am 'Add some fooBar')
4. Push to the branch (git push origin feature/fooBar)
5. Create a new Pull Request