PowerCurveResult#

class causalpy.experiments.sc_results.PowerCurveResult[source]#

Result of a simulation-based Bayesian power analysis.

Produced by SyntheticControl.power_analysis().

effect_sizes#

Candidate effect sizes evaluated.

Type:

list[float]

detection_rates#

Fraction of simulations where the criterion was met, per effect size.

Type:

list[float]

criterion#

The detection criterion used.

Type:

str

raw_results#

Nested list: per effect size, per simulation.

Type:

list[list[DressRehearsalResult]]

noise_method#

Residual-noise simulation method used.

Type:

str

block_length#

Block length used for block-bootstrap residual noise, if applicable.

Type:

int or None

Methods

PowerCurveResult.plot()

Power curve: effect size vs detection rate.

PowerCurveResult.summary()

DataFrame with per-effect-size summary statistics.

Attributes

block_length

noise_method

effect_sizes

detection_rates

criterion

raw_results

__init__(effect_sizes, detection_rates, criterion, raw_results, noise_method='iid_gaussian', block_length=None)#
Parameters:
Return type:

None

classmethod __new__(*args, **kwargs)#