---
id: bootstrapped
version: "0.0.2"
license: unclear
license_treatment: unclear
maintenance: abandoned
---

# bootstrapped — Implementations of the percentile based bootstrap

License: unclear · Maintenance: abandoned · Downloads: 88.0K/mo

## What it is and what it does

bootstrapped is a statistical library that computes confidence intervals by resampling data with replacement and analyzing the resulting distributions. It implements pivotal and percentile-based bootstrap methods to estimate error bars and bounds on population statistics from sample data. The library is designed for A/B testing and hypothesis validation, where you need to quantify uncertainty around observed metrics like conversion rates or mean values.

The package provides functions to calculate confidence intervals for single populations and comparative tests, plus statistical power analysis. It supports multithreaded computation and both dense and sparse arrays. However, the project has been abandoned since 2018-10-17, with no updates to handle modern Python or dependency versions, and license terms are not clearly specified in the package metadata.

Use it for:

- Compute error bars and confidence bounds for A/B test results to quantify improvement uncertainty
- Estimate population statistics from sample data with confidence intervals
- Perform statistical power analysis to determine required sample sizes for experiments
- Validate hypothesis test results by bootstrapping to understand sampling variation

## Worth the install?

AI-flagged interpretation of the facts above — verify before relying.

Builds confidence intervals from data samples using bootstrap resampling, with support for A/B testing and statistical power analysis.

No. The package is abandoned (last release 2018-10-17, repository archived) with no maintenance or security updates. High install friction from source-only distribution. License terms are unclear in metadata. Unless you are working with legacy code already using this package, choose an actively maintained alternative for bootstrap statistics.

## Install

pip install bootstrapped
uv add bootstrapped
poetry add bootstrapped

## Installing bootstrapped

Before you install: High install friction due to source-only distribution. Package is abandoned—last release was 2018-10-17 and repository was archived. No active maintenance or security updates.

License in practice: License treatment is unclear; the description mentions BSD licensing and patent grant, but the metadata does not specify SPDX or raw license text. Verify actual license terms before use.

Quickstart:

import bootstrapped.bootstrap as bs
import bootstrapped.stats_functions as bs_stats

samples = [1, 2, 3, 4, 5]
print(bs.bootstrap(samples, stat_func=bs_stats.mean))

Requires external dependencies (numpy, and matplotlib/pandas for power analysis). No Python version constraint specified in metadata.

Verify before relying:

- Whether the package works with current Python and dependency versions (last tested in 2018)
- Exact license terms and compatibility (BSD + patent grant specifics)
- Whether multithreading and sparse array support mentioned in description are production-ready

## Package facts

- License: not declared (unclear)
- Python support: unspecified
- Install friction: high
- Maintenance: abandoned
- Downloads: 88.0K/month (top 15,000 on PyPI)
- Known vulnerabilities: none known

## Tags

bootstrap confidence intervals, statistical resampling, a/b test analysis, error bars from samples, statistical power analysis, bootstrap statistics, population inference, statistics, bootstrap-resampling, abandoned

[View on SkillFed](https://skillfed.io/packages/bootstrapped) · [View on PyPI](https://pypi.org/project/bootstrapped/)