Metadata-Version: 2.4
Name: uee
Version: 0.1.0
Summary: Uniform Error Estimator (UEE): minimax / Chebyshev regression with diagnostics and bootstrap inference
Author: Philippe G. Penelle, Ph.D.
License-Expression: MIT
Project-URL: Homepage, https://github.com/PPenelle/uniform-error-estimator
Project-URL: Repository, https://github.com/PPenelle/uniform-error-estimator
Project-URL: Issues, https://github.com/PPenelle/uniform-error-estimator/issues
Project-URL: Changelog, https://github.com/PPenelle/uniform-error-estimator/blob/main/CHANGELOG.md
Keywords: econometrics,regression,minimax regression,chebyshev regression,uniform error estimator
Classifier: Development Status :: 3 - Alpha
Classifier: Intended Audience :: Science/Research
Classifier: Intended Audience :: Financial and Insurance Industry
Classifier: Programming Language :: Python :: 3
Classifier: Programming Language :: Python :: 3 :: Only
Classifier: Topic :: Scientific/Engineering :: Mathematics
Requires-Python: >=3.10
Description-Content-Type: text/markdown
License-File: LICENSE
Requires-Dist: numpy>=1.24
Requires-Dist: scipy>=1.11
Dynamic: license-file

# Uniform Error Estimator

**Uniform Error Estimator (UEE)** is a Python implementation of minimax (Chebyshev / L∞) regression:

\[
\\min\_eta \\max\_t |y\_t - X\_t eta|
]

* Pypi package: 'uee'
* Source repository: 'uniform-error-estimator'

The package provides:

* a clean estimator API (`UEE`)
* rich fit diagnostics (`UEEResults`)
* bootstrap inference
* optional intercept handling
* active-set and dual-weight diagnostics

## Installation

```bash
pip install uee
```

## Quick start

```python
import numpy as np
from uee import UEE

X = np.array(\\\[\\\[1.0], \\\[2.0], \\\[3.0], \\\[4.0]])
y = np.array(\\\[2.1, 4.0, 5.9, 8.2])

model = UEE(fit\\\_intercept=True)
res = model.fit(X, y)

print(res.intercept\\\_)
print(res.coef\\\_)
print(res.max\\\_abs\\\_error)
print(res.active\\\_set)
```

## Package status

This repository/package is the **maintained, packaged version** of UEE.
If you previously exposed a raw script or notebook implementation, update that legacy location to point users here.

## License

MIT. See `LICENSE`.

