MultiGroupCAMUV

class lingam.MultiGroupCAMUV(alpha=0.01, num_explanatory_vals=2, independence='hsic', ind_corr=0.5, prior_knowledge=None)

Implementation of CAM-UV Algorithm with multiple groups [1] [2] [3]

References

[1]

T.N.Maeda and S.Shimizu. Causal additive models with unobserved variables. In Proc. Thirty-Seventh Conference on Uncertainty in Artificial Intelligence (UAI). PMLR 161:97-106, 2021.

[2]

T. N. Maeda and S. Shimizu. Use of prior knowledge to discover causal additive models with unobserved variables and its application to time series data. Behaviormetrika, xx(xx): 1-19, 2024.

[3]

19. Shimizu. Joint estimation of linear non-Gaussian acyclic models. Neurocomputing, 81: 104-107, 2012.

__init__(alpha=0.01, num_explanatory_vals=2, independence='hsic', ind_corr=0.5, prior_knowledge=None)

Construct a CAM-UV model.

Parameters:

• alpha (float, optional (default=0.01)) – Alpha level.

• num_explanatory_vals (int, optional (default=2)) – Maximum number of explanatory variables.

• independence ({'hsic', 'fcorr'}, optional (default='hsic')) – Methods to determine independence. If ‘hsic’ is set, test for independence by HSIC. If ‘fcorr’ is set, independence is determined by F-correlation.

• ind_corr (float, optional (default=0.5)) – The threshold value for determining independence by F-correlation; independence is determined when the value of F-correlation is below this threshold value.

• prior_knowledge (array-like, shape ((index, index), ...), optional (default=None)) – List of variable pairs indicating prior knowledge. If (0, 3) is included , it means that X0 cannot be a cause of X3.

property adjacency_matrix_

Estimated adjacency matrix.

Returns:

adjacency_matrix_ – The adjacency matrix B of fitted model, where n_features is the number of features.

Return type:

array-like, shape (n_features, n_features)

fit(X_list)

Fit the model to multiple datasets.

Parameters:

X_list (list, shape [X, ...]) – Multiple datasets for training, where X is an dataset. The shape of ‘’X’’ is (n_samples, n_features), where n_samples is the number of samples and n_features is the number of features.

Returns:

self – Returns the instance itself.

Return type:

object