causalis.scenarios.unconfoundedness.refutation.score.residual_plots API Reference | Causalis

function

causalis.scenarios.unconfoundedness.refutation.score.residual_plots.plot_residual_diagnostics

plot_residual_diagnostics

Plot residual diagnostics for nuisance models.

Panels

Notes

These plots check the nuisance pieces directly:

outcome residuals $u_1 = Y - \hat g_1(X)$ and $u_0 = Y - \hat g_0(X)$ should look roughly centered around zero without strong patterns against fitted values,
treatment residuals $D - \hat m(X)$ should average near zero within propensity bins.

Clear trends usually mean the nuisance models still leave structure in the data, which can show up later as unstable score diagnostics.

Parameters

estimateCausalEstimate: Estimate with diagnostic data (m_hat, g0_hat; optionally g1_hat, y, d).
dataCausalData, optional: Optional fallback source for y and d when missing in diagnostic data.
clip_propensityfloat, default 1e-6: Clipping epsilon for propensity values in the treatment-residual panel.
n_binsint, default 20: Number of quantile bins for the binned-mean trend overlays.
marker_sizefloat, default 12.0: Scatter marker size.
alphafloat, default 0.35: Scatter opacity.
max_scatter_pointsint, optional: Maximum number of points drawn in each residual scatter panel. When set, scatter points are sampled uniformly without replacement, while the binned-mean overlays and calibration panel still use all observations.
random_stateint, optional: Random seed used when max_scatter_points triggers scatter sampling.
figsizetuple, default (14.0, 4.8): Figure size.
dpiint, default 220: Dots per inch.
font_scalefloat, default 1.10: Font scaling factor.
savestr, optional: Path to save the figure.
save_dpiint, optional: DPI for saving.
transparentbool, default False: Whether to save with transparency.

Returns

matplotlib.figure.Figure

The generated figure.

Examples

Canonical target

causalis.scenarios.unconfoundedness.refutation.score.residual_plots.plot_residual_diagnostics

Sections

PanelsNotesParametersReturnsExamples

data

causalis.scenarios.unconfoundedness.refutation.score.residual_plots.__all__

Value: ['plot_residual_diagnostics']

[‘plot_residual_diagnostics’]

Canonical target

causalis.scenarios.unconfoundedness.refutation.score.residual_plots.__all__