Extreme Joint Distributions

This library provides a C++ implementation of the Extreme Joint Distribution (EJD) algorithm in [1].

Dependencies

• CMake 3.15+
• C++17
• Blaze
• Boost 1.67+ (relies on Boost.Math)
• Discreture
• fmt
• ccache (optional)
• HighFive (optional for testing---reading of .mat files)

Build

The usual CMake build procedure

mkdir build
cd build
cmake .. -DCMAKE_BUILD_TYPE=Release
ninja -j4

References

If you find this library useful, please consider citing

@article{chiu2017correlated,
  title={Correlated multivariate Poisson processes and extreme measures},
  author={Chiu, Michael and Jackson, Kenneth R and Kreinin, Alexander},
  journal={Model Assisted Statistics and Applications},
  volume={12},
  number={4},
  pages={369--385},
  year={2017},
  publisher={IOS Press}
}

[1] Chiu, Michael, Kenneth R. Jackson, and Alexander Kreinin. "Correlated multivariate Poisson processes and extreme measures." Model Assisted Statistics and Applications 12.4 (2017): 369-385. [arxiv] [Model Assisted Statistics And Applications]