antiCPy 0.0.9.post3

Description:

antiCPy 0.0.9.post3

The antiCPy package provides tools to monitor destabilization because of varying control parameters or the influence of noise. Based on early warning measures it provides an extrapolation tool to estimate the time horizon in which a critical transition will probably occur.
antiCPy
The package abbreviation antiCPy stands for ''anticipate Critical Points (and if you like Change Points)
with Python''. The vision of the antiCPy package is designing a package collection of state-of-the-art
early warning measures, leading indicators and time series analysis tools that focus on system stability and
resilience in general as well as algorithms that might be helpful to estimate time horizons of future transitions or resilience changes.
It provides an easy applicable and efficient toolbox

to estimate the drift slope ζ^ of a polynomial Langevin equation as an early warning signal via Markov Chain Monte Carlo
(MCMC) sampling or maximum posterior (MAP) estimation,
to estimate a non-Markovian two-time scale polynomial system via MCMC or MAP with the option of a priori activated time scale separation,
to estimate the dominant eigenvalue by empiric dynamic modelling approaches like delay embedding and shadow manifolds combined with
iterated map's linear stability formalism,
to extrapolate an early warning signal trend to find the probable transition horizon based on the current data information.

Computationally expensive algorithms are implemented both, serially and strongly parallelized to minimize computation times. In case of
the change point trend extrapolation it involves furthermore algorithms that allow for computing of complicated fits with high numbers
of change points without memory errors.
The package aims to provide easily applicable methods and guarantee high flexibility and access to the derived interim results
for research purposes.

Citing antiCPy
If you use antiCPy's drift_slope measure, please cite
Martin Heßler et al. Bayesian on-line anticipation of critical transitions. New J. Phys. (2022). https://doi.org/10.1088/1367-2630/ac46d4.
If you use antiCPy's dominant_eigenvalue instead, please cite
Martin Heßler et al. Anticipation of Oligocene's climate heartbeat by simplified eigenvalue estimation.
arXiv (2023). https://doi.org/10.48550/arXiv.2309.14179
Documentation
You can find the documentation on read the docs.
Install
The package can be installed via
pip install antiCPy

Related publications
Up to now the package is accompanied by


the publication Efficient Multi-Change Point Analysis to Decode Economic Crisis Information from the S&P500 Mean Market Correlation,


the publication Memory Effects, Multiple Time Scales and Local Stability in Langevin Models of the S&P500 Market Correlation,


the publication Identifying dominant industrial sectors in market states of the S&P 500 financial data,


the publication Quantifying resilience and the risk of regime shifts under strong correlated noise,


the publication Bayesian on-line anticipation of critical transitions,


the preprint Anticipation of Oligocene's climate heartbeat by simplified eigenvalue estimation,


the preprint Quantifying Tipping Risks in Power Grids and beyond.