Software

pycop  Python Package

PyPi version Downloads

pycop is a Python package for dependence modeling with copulas. It is currently one of the most comprehensive tools for modelling dependence in Python and is regularly used by quantitative analysts and researchers worldwide.

The package was developed to address limitations in existing Python tools when modeling complex multivariate dependence structures, particularly in the context of financial risk analysis and extreme value theory.

Features
· Archimedean copulas (Clayton, Gumbel, Frank, Joe…)
· Elliptical copulas (Gaussian, Student-t)
· Tail dependence coefficient estimation
· Simulation & calibration tools
· Empirical copula estimation
· Visualization utilities
Applications
· Portfolio risk modeling
· Joint tail risk estimation
· Contagion analysis
· Stress testing
· Extreme value dependence
· Replication studies
Cite as: Nicolas, M. L. (2022). pycop: a Python package for dependence modeling with copulas. Zenodo Software Package. doi:10.5281/zenodo.7030034
Links:  Zenodo (DOI)  ·  GitHub (@maximenc)

Installation

# Install via pip
pip install pycop

# Basic usage
from pycop import simulation, archimedean
cop = archimedean.assoc("clayton", dim=2)
sample = simulation.simu_archimedean("clayton", n=1000, theta=2)

Related Publications

[1]
Nicolas, M. L. (2025). Tail risk exposure and the cross section of expected stock returns. Journal of Banking & Finance. Uses pycop for tail dependence estimation.
ssrn.com/abstract=4956412
[2]
Garcin, M., & Nicolas, M. L. (2024). Nonparametric estimator of the tail dependence coefficient: balancing bias and variance. Statistical Papers. Benchmark study using copula simulation framework.
10.48550/arXiv.2111.11128