data
causalis.shared.confounder_clustering.CorrelationMethodCorrelationMethod
Value: None
None
Python Docs
causalis
Reference
Package entry
causalisRoot package overview and namespace mapNamespaces
Submodule
causalis.shared.confounder_clusteringconfounder_clustering
Submodule causalis.shared.confounder_clustering with no child pages and 6 documented members.
Functions
Jump directly into the documented functions for this page.
2 items
Data
Jump directly into the documented data for this page.
4 items
data
causalis.shared.confounder_clustering.LinkageMethodLinkageMethod
Value: None
None
data
causalis.shared.confounder_clustering.OutcomeImportanceMethodOutcomeImportanceMethod
Value: None
None
function
causalis.shared.confounder_clustering.cluster_confounderscluster_confounders
Cluster confounders using their absolute pairwise correlations.
The clustering distance is
d(Xi,Xj)=1−∣corr(Xi,Xj)∣.
Hierarchical clustering is cut at 1 - min_abs_correlation. Negative
and positive correlations of the same magnitude are therefore treated as
equally similar. By default, complete linkage is used, so every pair of
features in a non-singleton cluster has an absolute correlation at least
as large as min_abs_correlation on the sampled data.
Parameters
- dataCausalData or MultiCausalData
Causal data contract exposing numeric
dfandconfoundersattributes.- min_abs_correlationfloat, default 0.7
Correlation cut-off in the closed interval [0, 1]. Larger values form smaller, more strongly related clusters.
- correlation_method{“pearson”, “spearman”}, default “pearson”
Pairwise correlation measure. Spearman correlation is useful for monotonic non-linear relationships but requires more work and memory.
- linkage_method{“average”, “complete”, “single”}, default “complete”
Hierarchical-linkage rule. Only complete linkage guarantees the pairwise threshold interpretation described above.
- max_samplesint or None, default 200000
Maximum number of rows used to estimate correlations. If the dataset is larger, rows are sampled without replacement. Pass
Noneto use all rows.- random_stateint or None, default 42
Seed used for row sampling.
Returns
list[list[str]]
Clusters in the original confounder order. Features within each cluster also preserve their original order. Singleton clusters are retained.
Examples
Notes
Missing correlations can occur when a rare feature is constant in the sampled rows. They are treated as zero correlation, leaving that feature separate unless it is linked to another feature by valid correlations.
Canonical target
causalis.shared.confounder_clustering.cluster_confounders
Sections
ParametersReturnsNotesExamples
function
causalis.shared.confounder_clustering.rank_confounder_clustersrank_confounder_clusters
Cluster confounders and rank the clusters by joint nuisance importance.
For each correlation cluster Cg, the treatment and outcome importances are aggregated as
ID,g=j∈Cg∑Im,j,
and, by default,
IY,g=21j∈Cg∑Ig0,j+j∈Cg∑Ig1,j.
Clusters are sorted by score = importance_d * importance_y in
descending order. Native importance vectors are normalized separately for
m, g0, and g1 before aggregation.
Parameters
- dataCausalData or MultiCausalData
Data containing the confounders to cluster.
- effect_estimationAny
Fitted IRM, causal estimate with diagnostic data, or dictionary containing one of them. Feature importance must have been collected. min_abs_correlation, correlation_method, linkage_method, max_samples, random_state Forwarded to :func:
cluster_confounders.- outcome_importance{“mean”, “max”}, default “mean”
Combine the cluster-level
g0andg1importance by their mean or maximum."max"is more conservative for features important in only one treatment arm.
Returns
pandas.DataFrame
Cluster ranking with columns cluster_id, features, n_features, importance_d, importance_g0, importance_g1, importance_y, and score. Row zero is the highest-ranked cluster; features can be passed directly to sensitivity_benchmark_group.
Examples
Canonical target
causalis.shared.confounder_clustering.rank_confounder_clusters
Sections
ParametersReturnsExamples
data
causalis.shared.confounder_clustering.__all____all__
Value: ['cluster_confounders', 'rank_confounder_clusters']
[‘cluster_confounders’, ‘rank_confounder_clusters’]