Skip to main content

Non-Linear Shrinkage

Provides a function that calculates an estimate of the covariance matrix shrunk using a non-linear analytic formula provided by the working paper Ledoit and Wolf (2018), entitled ['Analytical Nonlinear Shrinkage of Large-Dimensional Covariance Matrices'] (http://www.econ.uzh.ch/static/wp/econwp264.pdf).

Installation

pip install nonlinshrink

Usage

import numpy as np
import nonlinshrink as nls
p = 2
n = 13
sigma = np.eye(p, p)
data = np.random.multivariate_normal(np.zeros(p), sigma, n)
sigma_tilde = nls.shrink_cov(data)

The data is automatically demeaned unless otherwise specified. In the case where the data has been pre-processed and the effective degrees of freedom of the dataset is decreased, e.g. through an OLS regression, the user can specify this through a parameter k which signifies the degrees of freedom already subtracted. For example,

import numpy as np
import nonlinshrink as nls
p = 2
n = 14
sigma = np.eye(p, p)
data = np.random.multivariate_normal(np.zeros(p), sigma, n) + np.arange(n)[:, np.newaxis] + 1
x = np.vstack((np.ones(n).T, np.arange(n).T)).T
betahat = np.linalg.solve(np.dot(x.T, x), np.dot(x.T, data))
datahat = np.dot(x, betahat)
res = data - datahat
sigma_tilde = nls.shrink_cov(res, k=2)  # corresponding to 2 degrees of freedom

Developing

Please submit a PR! The shrinkage function itself is located in nonlinshrink.py. For running the tests do

git clone https://github.com/matzhaugen/analytic_shrinkage.git
cd analytic_shrinkage
pip install -e . # install the package
pytest

Metadata

Release files for nonlinshrink 0.7

For a detailed explanation of source distributions (sdists) and built distributions (wheels), please see the package formats documentation.

Built distribution (wheel)

Table of built distributions (wheels) for nonlinshrink 0.7
File Interpreter ABI Platform Actions
nonlinshrink-0.7-py3-none-any.whl Python 3 none any Details for nonlinshrink-0.7-py3-none-any.whl

Release files / nonlinshrink-0.7-py3-none-any.whl

Download URL nonlinshrink-0.7-py3-none-any.whl
Size 6.3 kB
Tags Python 3
SHA-256 checksum
How to use checksums
3bbc13ddc9283993dc08f89b90f160c3b389aa30f12d55282436bf9785e2471c
BLAKE2b-256 checksum
How to use checksums
8b0aaeb20552a2db5cfcc9e1c5b831295ef9785ea2eb9d428dce2d23d16c0da3
Upload date
Uploaded using Trusted Publishing?
What is trusted publishing?
No
Uploaded via twine/2.0.0 pkginfo/1.5.0.1 requests/2.21.0 setuptools/41.6.0 requests-toolbelt/0.9.1 tqdm/4.19.1 CPython/3.7.3

Security for nonlinshrink 0.7

Provenance

Provenance describes where a file came from. On PyPI, provenance is shared via attestations, which provide a verifiable record of the build or publishing details. View details, limitations and caveats.

No provenance found

None of the files in this release were uploaded with provenance metadata.

A note on package security

PyPI is a repository for distributing packages, not a service for auditing them. While third-party security researchers and our staff do their best to detect and remove malicious packages, PyPI cannot guarantee the quality, security or suitability of any given package.

To learn more about protecting yourself from malicious actors, visit our package security documentation.

Release history Release notifications | RSS feed

This release

0.7 This release

1 release file No provenance

Anthropic, PBC Visionary sponsor Bloomberg Visionary sponsor Hudson River Trading Visionary sponsor Meta Visionary sponsor NVIDIA Visionary sponsor Microsoft Sustainability sponsor Depot Continuous Integration AWS Cloud computing and Security Sponsor Datadog Monitoring Fastly CDN Google Download Analytics Sentry Error logging StatusPage Status page