SCSPILLResults

utils.scspill_helpers.structures.SCSPILLResults()

Standardized results of the SCSPILL estimator.

Subclasses :class:scspill.config_models.BaseEstimatorResults, so the flat effect surface (att, att_ci, counterfactual, gap, donor_weights, pre_rmse, .plot()) resolves through the shared contract, while the Bayesian detail lives in typed fields:

• :attr:inputs – the prepared panel and spatial weights;
• :attr:alpha_posterior / :attr:sar_posterior – the two posterior blocks with their full draw arrays;
• :attr:effects_detail – ATT draws, counterfactual band, and the per-donor spillover panel;
• :attr:scm_weights – classical simplex-SCM comparator weights.

:attr:method records which spillover model produced the fit. It mirrors SCSPILLConfig.method and the suffix of method_details.method_name ("SCSPILL/<method>"). Only "sar" exists today; a model added later declares its own posterior blocks alongside :attr:alpha_posterior / :attr:sar_posterior, which are SAR-specific, while :attr:inputs, :attr:effects_detail and the flat effect surface are the cross-model contract.

Attributes

Name	Description
acc_rho	Post-burn acceptance rate of the rho Metropolis step.
alpha_draws	Step-1 alpha draws, shape (M, N).
alpha_hat	Posterior mean of the synthetic weights alpha.
rho_ci	Equal-tailed credible interval for rho at the fit’s ci level.
rho_draws	Step-2 rho draws, shape (M,).
rho_ess	Effective sample size of the rho chain.
rho_hat	Posterior mean of the spillover intensity rho.
spillover_lower	Pointwise lower credible band of the spillover effects.
spillover_panel	Posterior-mean spillover effects, (T, N) time-by-donor.
spillover_upper	Pointwise upper credible band of the spillover effects.

Methods

Name	Description
diagnostics	Posterior summary and convergence diagnostics table.
plot	Plot the fitted result.

diagnostics

utils.scspill_helpers.structures.SCSPILLResults.diagnostics(
    which_alpha=None,
    top_n_alpha=6,
    which_beta=None,
    top_n_beta=6,
)

Posterior summary and convergence diagnostics table.

Builds one row per monitored chain (rho, sigma2, selected alpha components, and covariate coefficients when present) with posterior mean, sd, quantiles, effective sample size, split-chain R-hat, Monte Carlo standard error, and a Geweke z-score – mirroring the R package’s diagnostics() summary.

Parameters

Name	Type	Description	Default
which_alpha	list	Donor labels whose alpha chains to include. Defaults to the top_n_alpha donors by absolute posterior-mean weight.	None
top_n_alpha	int	Number of top-|alpha| donors monitored when which_alpha is None.	6
which_beta	list	Covariate names whose beta chains to include. Defaults to the first top_n_beta covariates.	None
top_n_beta	int	Number of covariates monitored when which_beta is None.	6

Returns

Name	Type	Description
	pd.DataFrame	One row per parameter with columns mean, sd, q025, q50, q975, ess, rhat_split, mcse, geweke_z.

plot

utils.scspill_helpers.structures.SCSPILLResults.plot(
    kind='auto',
    *,
    ax=None,
    **overrides,
)

Plot the fitted result.

Parameters

Name	Type	Description	Default
kind	str	"auto" / "counterfactual" / "gap" use the shared effect-plot contract; "panel", "full", "effect", "spill_top", "weights", "rho", and "trace" route to :func:scspill.utils.scspill_helpers.plotter.plot_scspill.	"auto"
ax	matplotlib Axes	Draw into an existing axis (single-panel kinds only).	None
**overrides	Any	Cosmetic overrides forwarded to the plotting layer.	{}

Returns

Name	Type	Description
	matplotlib.axes.Axes or np.ndarray of Axes