bootstrap_analysis

bootstrap_analysis(
    data,
    distribution,
    probabilities,
    estimation_method='MaximumLikelihood',
    sample_size=None,
    replications=1000,
    seed=12345,
    alpha=0.1,
)

Parametric bootstrap confidence bands for a fitted distribution.

Fits distribution to data, then resamples it replications times to derive percentile confidence bands on the quantile curve at the non-exceedance probabilities. Wraps the shared C++ BootstrapAnalysis (Numerics).

Parameters

Name	Type	Description	Default
data	array_like	Observations to fit.	required
distribution	str	Distribution family name.	required
probabilities	array_like of float	Non-exceedance probabilities at which to evaluate the quantile curve.	required
estimation_method	str	Estimation method used to fit the distribution.	"MaximumLikelihood"
sample_size	int	Bootstrap sample size; when omitted, uses len(data).	None
replications	int	Number of bootstrap replications.	1000
seed	int	PRNG seed for the resampling.	12345
alpha	float	Significance level of the confidence bands.	0.1

Returns

Name	Type	Description
	dict	Percentile confidence bands on the quantile curve at probabilities.