Skip to content
Submodule
causalis.scenarios.unconfoundedness.refutation.overlap.overlap_validation

overlap_validation

Submodule causalis.scenarios.unconfoundedness.refutation.overlap.overlap_validation with no child pages and 2 documented members.

Symbol indexAPI membersrun_overlap_diagnostics__all__

Functions

Jump directly into the documented functions for this page.

1 items
run_overlap_diagnosticsfunction

Data

Jump directly into the documented data for this page.

1 items
__all__data
function
causalis.scenarios.unconfoundedness.refutation.overlap.overlap_validation.run_overlap_diagnostics

run_overlap_diagnostics

Run overlap and calibration diagnostics for an estimated propensity model.

The core overlap object is the propensity score

m(X)=P(D=1∣X).m(X) = \mathbb{P}(D=1 \mid X).

This diagnostic checks whether estimated propensities stay away from the edges and whether the implied weights are stable. For example, ATE weights use

wi(1)=Dim(Xi),wi(0)=1−Di1−m(Xi),w_i^{(1)} = \frac{D_i}{m(X_i)}, \qquad w_i^{(0)} = \frac{1-D_i}{1-m(X_i)},

so very small m(Xi)m(X_i) or very large m(Xi)m(X_i) can create large leverage points. The report combines:

  • edge mass near 0 and 1,

  • treated/control separation in propensity space (KS, AUC),

  • empirical common support on the logit propensity scale,

  • effective sample size and tail diagnostics for weights,

  • calibration summaries such as ECE, recalibration slope, and intercept.

The support_rate metric describes empirical common support in the observed sample. It does not prove the theoretical positivity assumption and does not test for unobserved confounding.

Parameters

dataCausalData

Dataset used to fit the estimator.

estimateCausalEstimate

Effect estimate with diagnostic_data containing propensity-related arrays such as m_hat and d.

thresholdsdict, optional

Optional threshold overrides keyed by metric name.

n_binsint, default 10

Number of bins used for calibration summaries.

use_hajekbool, optional

Whether to evaluate normalized IPW identities. If omitted, the value is inferred from diagnostic metadata.

return_summarybool, default True

Include a compact tabular summary in the returned payload.

auc_flip_marginfloat, default 0.05

Margin around 0.5 used when flagging reversed treated/control ranking.

Returns

Dict[str, Any]

Diagnostic report containing edge-mass, calibration, weight-stability, and optional summary tables.

Raises

ValueError

If required diagnostic arrays are missing or have incompatible shapes.

Examples

Canonical target

causalis.scenarios.unconfoundedness.refutation.overlap.overlap_validation.run_overlap_diagnostics

Sections

ParametersReturnsRaisesExamples
Link to this symbol
data
causalis.scenarios.unconfoundedness.refutation.overlap.overlap_validation.__all__

__all__

Value: ['run_overlap_diagnostics']

[‘run_overlap_diagnostics’]

Canonical target

causalis.scenarios.unconfoundedness.refutation.overlap.overlap_validation.__all__

Link to this symbol