causalis.data_contracts.rct_estimates

Mapping of RCT outcome/arm estimates with a combined summary table.

Module Contents

Classes

RctEstimates

Batch estimates, indexed by outcome and then active treatment arm.

API

class causalis.data_contracts.rct_estimates.RctEstimates

Bases: dict[str, dict[str, causalis.data_contracts.causal_estimate.CausalEstimate]]

Batch estimates, indexed by outcome and then active treatment arm.

Retains ordinary dictionary access, iteration, and items(). summary combines the existing CausalEstimate summaries into a notebook-friendly table with one column per treatment-versus-control comparison. Values and confidence intervals use exactly the single-estimate formatting.

Initialization

Initialize self. See help(type(self)) for accurate signature.

summary(outcome: str | None = None, treatment: str | None = None) → pandas.DataFrame

Summarize one outcome or all outcomes, optionally selecting an arm.

Rows are the fields from CausalEstimate.summary(). When outcome is specified, columns are comparison labels, e.g. variant_a vs control. Otherwise columns have two levels: outcome and comparison. Binary treatment comparisons use <treatment>=1 vs <treatment>=0. Unknown selectors raise ValueError. The returned table is independent of the stored estimates; no model fitting or inference is repeated.