Bootstrap aggregation and confidence measures to improve time series causal discovery

In this folder, you can find the official implementation of the paper:

Debeire, K., Runge, J., Gerhardus, A., Eyring, V. (2023). Bootstrap aggregation and confidence measures to improve time series causal discovery (see arXiv article)

A method to perform bootstrap aggregation of time series causal graphs and measure a confidence for links of the aggregated graph. Here combined with the PCMCI+ algorithm from the TIGRAMITE package. We provide here the code that can be used to recreate figures 3, 4, and 5 as presented in the paper.

Author: Kevin Debeire, [email protected]

The current release on Zenodo can be found here:

Requirements

First setup a conda environment (by default called bagged_pcmci) from the environment.yml file:

conda env create -f environment.yml

Activate this environment.

As explained in the paper, our bagging approach and confidence measures is combined with the PCMCI+ causal discovery algorithm. The PCMCI+ method is implemented in the TIGRAMITE package. Follow the instructions below to install TIGRAMITE.

First clone the TIGRAMITE repository:

git clone https://github.com/jakobrunge/tigramite.git

and point to this specific commit

git reset --hard 27ded041e87f8d9a8d5f8714e8db5c1235e8616a

Then, our bagging and confidence measures method is implemented in the scripts provided in the folder to_replace_in_tigramite/.

You will have to replace the TIGRAMITE pcmci_base.py and data_processing.py files in tigramite/tigramite with the ones provided in to_replace_in_tigramite/, respectively pcmci_base.py and data_processing.py.

The modified pcmci_base.py and data_processing.py include the bagging and confidence measures introduced in the paper.

Then install TIGRAMITE:

python setup.py install

You should now be able to run the numerical experiments and reproduce the figures of the paper.

Generating numerical experiments data and plotting figures

Find below the instructions to produce the figure 3, 4, 5A and 5B of the paper. All default model and method parameters of the scripts are already set to reproduce the figures.

Please adjust the save paths (by default ./ ) in the following scripts : compute_fig4and3.py, compute_fig5A.py, compute_fig5B.py, compute_metrics_fig5B.py, plotfig4and3.py, plot_fig5A.py, plot_fig5B.py. They specify where the numerical results are saved, and where the figures are saved.

For figure 3 and figure 4:

• If you have access to an HPC system with slurm job scheduler:
  • fill in sbatch_fig4and3.sh. Specify your account, partition, number of cores per CPU, etc...
  • check the model parameters in create_submission_fig4and3.py
  • run: 'python create_submission_fig4and3.py 1'. This will submit a job for each model configurations (here for each alpha_pc of PCMCI+) in which the Bagged-PCMCI+ and PCMCI+ are evaluated on the generated synthetic data
  • Once all computations are done. Run 'python plot_fig4and3.py par_corr pc_alpha_highdegree'
  • You can modify the values of pc_alpha in 'create_submission_fig4and3.py' and 'plot_fig4and3' to produce figure 4
• Alternatively, it can be run locally, but this is not recommended as the computational cost to run the numerical experiments are high:
  • set run_locally=True and submit=False in create_submission_fig4and3.py
  • run: 'python create_submission_fig4and3.py 1'
  • Once all computations are done, run 'python plot_fig4and3.py par_corr pc_alpha_highdegree'.

For figure 5A:

• If you have access to an HPC system with slurm job scheduler:

  • fill in sbatch_fig5A.sh. Specify your account, partition, number of cores per CPU, etc...
  • check the model parameters in compute_fig5A.py
  • run: 'sbatch sbatch_fig5A.sh'. This will submit a job for the specified method and model settings. Bagged-PCMCI+ and PCMCI+ are evaluated on generated synthetic data
  • Once all computations are done, run 'python plot_fig5A.py' to reproduce the figure 5A.

• Alternatively, it can be run locally, but this is not recommended as the computational cost to run the numerical experiments are high:

  • check the model parameters in compute_fig5A.py
  • run: 'python compute_fig5A.py' to generate the synthetic data and run Bagged-PCMCI+ and PCMCI+
  • Once all computations are done, run 'python plot_fig5A.py' to reproduce the figure 5A.

For figure 5B:

• If you have access to an HPC system with slurm job scheduler:

  • fill in sbatch_fig5B.sh. Specify your account, partition, number of cores per CPU, etc...
  • check the model parameters in create_submission_fig5B.py
  • run: 'python create_submission_fig5B.py 1'. This will submit a job for the specified method and model settings
  • run: 'python compute_metrics_fig5B.py'. This will calculate the mean absolute frequencies errors
  • Run 'python plot_fig5B.py' to reproduce the figure 5A.

• Alternatively, it can be run locally, but this is not recommended as the computational cost to run the numerical experiments are high:

  • set run_locally=True and submit=False in create_submission_fig5B.py
  • run: 'python create_submission_fig5B.py 1'
  • Once all computations are done, run: 'python compute_metrics_fig5B.py'. This will calculate the mean absolute frequencies errors
  • Run 'python plot_fig5B.py par_corr boot_rea_highdegree' to reproduce the figure 5B.

License

GNU General Public License v3.0

See license.txt for full text.