causalcompass.algorithms.VAR

class causalcompass.algorithms.VAR(tau_max=3, threshold=0.01, seed=None, **kwargs)[source]

Granger causality-based causal discovery method that fits a Vector Autoregressive model and infers causal relationships from the estimated coefficient matrix.

References

https://github.com/cloud36/graphical_granger_methods

Parameters:

  • tau_max (int, default 3) – Maximum time lag

  • threshold (float, default 0.01) – Coefficient threshold for edge filtering

Examples

>>> from causalcompass.algorithms import VAR
>>> model = VAR(tau_max=3, threshold=0.01)
>>> predicted_adj = model.run(X)
>>> all_metrics, no_diag_metrics = model.eval(true_adj, predicted_adj)
__init__(tau_max=3, threshold=0.01, seed=None, **kwargs)[source]

Initialize VAR

Methods

__init__([tau_max, threshold, seed])

Initialize VAR

eval(true_adj, predicted_adj[, shd_thresholds])

Evaluate the predicted adjacency matrix against the ground truth.

run(X)

Run VAR algorithm.

run_raw(X)

Run the algorithm and return an unthresholded intermediate result that can be reused across multiple threshold values.

run_threshold_sweep(X, thresholds)

Run the algorithm once and post-process the raw result for each threshold.