bootstrapx-lib 0.2.0

Production-grade uncertainty estimation for Python.

bootstrapx

Production-grade uncertainty estimation for Python.

14 bootstrap methods · Numba JIT · Optional CUDA GPU · Memory-safe batching

---

Why bootstrapx?

scipy.stats.bootstrap supports only 3 CI methods and has no time-series support. The R boot package is comprehensive but not accessible from Python. bootstrapx bridges this gap with 14 methods, Numba acceleration, and a clean API.

Feature	scipy	R boot	bootstrapx
BCa interval	✅	✅	✅
Studentized (bootstrap-t)	❌	✅	✅
Poisson / Bernoulli weights	❌	❌	✅
Time-series (MBB, CBB, Stationary, Sieve, Tapered, Wild)	❌	Partial	✅
Cluster / Stratified	❌	Partial	✅
Numba JIT	❌	N/A	✅
CUDA GPU	❌	❌	✅
Generator batching (constant memory)	❌	❌	✅

---

Installation

pip install bootstrapx-lib

# With GPU
pip install "bootstrapx-lib[cuda]"

---

Quick Start

import numpy as np
from bootstrapx import bootstrap

data = np.random.default_rng(42).normal(5, 2, size=200)

result = bootstrap(data, np.mean)
print(result)
# BootstrapResult(method='bca', theta_hat=4.94, se=0.13, CI=[4.70, 5.19])

---

Performance

bootstrapx is significantly faster than scipy.stats.bootstrap for large datasets:

N	Method	Scipy	Bootstrapx	Speedup
5,000	BCa	0.80s	0.27s	3.0x
50,000	Percentile	7.29s	2.01s	3.6x
100,000	Percentile	54.34s	3.99s	13.6x

Benchmark on Apple M1, Python 3.12. See Benchmarks for details.

---

Coverage Accuracy

Monte Carlo simulation ($N=1000$, 500 runs) confirms correct statistical coverage:

Distribution	BCa Coverage (Nominal 95%)
Normal	94.8%
Skewed (Exponential)	95.0%
Heavy-Tailed (t-dist)	94.0%

---

Documentation

📖 Full docs: artyerokhin.github.io/bootstrapx

• Getting Started
• Methods Guide — math behind each method
• API Reference
• Benchmarks

---

Contributing

git clone https://github.com/artyerokhin/bootstrapx.git
cd bootstrapx
pip install -e ".[dev,docs]"
pytest tests/ -v
mkdocs serve

---

License

MIT — see LICENSE.