HierarchicalInterruptedTimeSeries.plot#

HierarchicalInterruptedTimeSeries.plot(*, ci_prob=0.94, show=True, legend_kwargs=None)[source]#

Plot the fitted hierarchical model’s population-level effect.

Dispatches on effect_type: a forest plot of per-unit lifts plus the population posterior for "instant", the population Hill saturation curve for "saturation", or an event-study plot of per-bin population effects for "event_study" / "placebo". See plot_unit() to plot a single unit’s observed vs. counterfactual trajectory instead.

Parameters:

  • ci_prob (float) – Probability mass of the HDI band drawn around per-unit lifts, the saturation curve, or per-bin event-study effects. Must be in (0, 1]. Defaults to HDI_PROB.

  • show (bool) – Whether to automatically display the plot. Defaults to True.

  • legend_kwargs (dict[str, Any] | None) – Keyword arguments to adjust legend placement and styling. Supported keys: loc, bbox_to_anchor, fontsize, frameon, title (bbox_transform is accepted alongside bbox_to_anchor).

Returns:

Matplotlib figure and axes (a pair of axes for "instant", a single axes otherwise).

Return type:

fig, ax