| Name | skfolio JSON |
| Version |
0.11.0
JSON |
| download |
| home_page | None |
| Summary | Portfolio optimization built on top of scikit-learn |
| upload_time | 2025-07-26 20:36:29 |
| maintainer | None |
| docs_url | None |
| author | None |
| requires_python | >=3.10 |
| license | BSD 3-Clause License
Copyright (c) 2007-2023 The skfolio developers.
All rights reserved.
Redistribution and use in source and binary forms, with or without
modification, are permitted provided that the following conditions are met:
* Redistributions of source code must retain the above copyright notice, this
list of conditions and the following disclaimer.
* Redistributions in binary form must reproduce the above copyright notice,
this list of conditions and the following disclaimer in the documentation
and/or other materials provided with the distribution.
* Neither the name of the copyright holder nor the names of its
contributors may be used to endorse or promote products derived from
this software without specific prior written permission.
THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS "AS IS"
AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE
DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE
FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR
SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER
CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE
OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE. |
| keywords |
portfolio
optimization
optimisation
finance
asset
allocation
quantitative
quant
investment
strategy
machine-learning
scikit-learn
data-mining
data-science
|
| VCS |
 |
| bugtrack_url |
|
| requirements |
No requirements were recorded.
|
| Travis-CI |
No Travis.
|
| coveralls test coverage |
No coveralls.
|
.. -*- mode: rst -*-
|Licence| |Codecov| |Black| |PythonVersion| |PyPi| |CI/CD| |Downloads| |Ruff| |Contribution| |Website| |JupyterLite| |Discord| |DOI|
.. |Licence| image:: https://img.shields.io/badge/License-BSD%203--Clause-blue.svg
:target: https://github.com/skfolio/skfolio/blob/main/LICENSE
.. |Codecov| image:: https://codecov.io/gh/skfolio/skfolio/graph/badge.svg?token=KJ0SE4LHPV
:target: https://codecov.io/gh/skfolio/skfolio
.. |PythonVersion| image:: https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-blue.svg
:target: https://pypi.org/project/skfolio/
.. |PyPi| image:: https://img.shields.io/pypi/v/skfolio
:target: https://pypi.org/project/skfolio
.. |Black| image:: https://img.shields.io/badge/code%20style-black-000000.svg
:target: https://github.com/psf/black
.. |CI/CD| image:: https://img.shields.io/github/actions/workflow/status/skfolio/skfolio/release.yml.svg?logo=github
:target: https://github.com/skfolio/skfolio/raw/main/LICENSE
.. |Downloads| image:: https://static.pepy.tech/badge/skfolio
:target: https://pepy.tech/project/skfolio
.. |Ruff| image:: https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json
:target: https://github.com/astral-sh/ruff
.. |Contribution| image:: https://img.shields.io/badge/Contributions-Welcome-blue
:target: https://github.com/skfolio/skfolio/blob/main/CONTRIBUTING.md
.. |Website| image:: https://img.shields.io/website.svg?down_color=red&down_message=down&up_color=53cc0d&up_message=up&url=https://skfolio.org
:target: https://skfolio.org
.. |JupyterLite| image:: https://jupyterlite.rtfd.io/en/latest/_static/badge.svg
:target: https://skfolio.org/lite
.. |Discord| image:: https://img.shields.io/badge/Discord-Join%20Chat-5865F2?logo=discord&logoColor=white
:target: https://discord.gg/Bu7EtNYugS
.. |DOI| image:: https://zenodo.org/badge/731792488.svg
:target: https://doi.org/10.5281/zenodo.16148630
.. |PythonMinVersion| replace:: 3.10
.. |NumpyMinVersion| replace:: 1.23.4
.. |ScipyMinVersion| replace:: 1.8.0
.. |PandasMinVersion| replace:: 1.4.1
.. |CvxpyBaseMinVersion| replace:: 1.5.0
.. |ClarabelMinVersion| replace:: 0.9.0
.. |SklearnMinVersion| replace:: 1.6.0
.. |JoblibMinVersion| replace:: 1.3.2
.. |PlotlyMinVersion| replace:: 5.22.0
===============
|icon| skfolio
===============
.. |icon| image:: https://raw.githubusercontent.com/skfolio/skfolio/master/docs/_static/logo_animate.svg
:width: 100
:alt: skfolio documentation
:target: https://skfolio.org/
**skfolio** is a Python library for portfolio optimization and risk management built on
top of scikit-learn. It offers a unified interface and tools compatible with
scikit-learn to build, fine-tune, cross-validate and stress-test portfolio models.
It is distributed under the open-source 3-Clause BSD license.
.. image:: https://raw.githubusercontent.com/skfolio/skfolio/master/docs/_static/expo.jpg
:target: https://skfolio.org/auto_examples/
:alt: examples
Important links
~~~~~~~~~~~~~~~
- Documentation: https://skfolio.org/
- Examples: https://skfolio.org/auto_examples/
- User Guide: https://skfolio.org/user_guide/
- GitHub Repo: https://github.com/skfolio/skfolio/
Installation
~~~~~~~~~~~~
`skfolio` is available on PyPI and can be installed with::
pip install -U skfolio
Dependencies
~~~~~~~~~~~~
`skfolio` requires:
- python (>= |PythonMinVersion|)
- numpy (>= |NumpyMinVersion|)
- scipy (>= |ScipyMinVersion|)
- pandas (>= |PandasMinVersion|)
- cvxpy-base (>= |CvxpyBaseMinVersion|)
- clarabel (>= |ClarabelMinVersion|)
- scikit-learn (>= |SklearnMinVersion|)
- joblib (>= |JoblibMinVersion|)
- plotly (>= |PlotlyMinVersion|)
Docker
~~~~~~
You can also spin up a reproducible JupyterLab environment using Docker:
Build the image::
docker build -t skfolio-jupyterlab .
Run the container::
docker run -p 8888:8888 -v <path-to-your-folder-containing-data>:/app/data -it skfolio-jupyterlab
Browse:
Open localhost:8888/lab and start using `skfolio`
Key Concepts
~~~~~~~~~~~~
Since the development of modern portfolio theory by Markowitz (1952), mean-variance
optimization (MVO) has received considerable attention.
Unfortunately, it faces a number of shortcomings, including high sensitivity to the
input parameters (expected returns and covariance), weight concentration, high turnover,
and poor out-of-sample performance.
It is well-known that naive allocation (1/N, inverse-vol, etc.) tends to outperform
MVO out-of-sample (DeMiguel, 2007).
Numerous approaches have been developed to alleviate these shortcomings (shrinkage,
additional constraints, regularization, uncertainty set, higher moments, Bayesian
approaches, coherent risk measures, left-tail risk optimization, distributionally robust
optimization, factor model, risk-parity, hierarchical clustering, ensemble methods,
pre-selection, etc.).
Given the large number of methods, and the fact that they can be combined, there is a
need for a unified framework with a machine-learning approach to perform model
selection, validation, and parameter tuning while mitigating the risk of data leakage
and overfitting.
This framework is built on scikit-learn's API.
Available models
~~~~~~~~~~~~~~~~
* Portfolio Optimization:
* Naive:
* Equal-Weighted
* Inverse-Volatility
* Random (Dirichlet)
* Convex:
* Mean-Risk
* Risk Budgeting
* Maximum Diversification
* Distributionally Robust CVaR
* Clustering:
* Hierarchical Risk Parity
* Hierarchical Equal Risk Contribution
* Nested Clusters Optimization
* Ensemble Methods:
* Stacking Optimization
* Expected Returns Estimator:
* Empirical
* Exponentially Weighted
* Equilibrium
* Shrinkage
* Covariance Estimator:
* Empirical
* Gerber
* Denoising
* Detoning
* Exponentially Weighted
* Ledoit-Wolf
* Oracle Approximating Shrinkage
* Shrunk Covariance
* Graphical Lasso CV
* Implied Covariance
* Distance Estimator:
* Pearson Distance
* Kendall Distance
* Spearman Distance
* Covariance Distance (based on any of the above covariance estimators)
* Distance Correlation
* Variation of Information
* Distribution Estimator:
* Univariate:
* Gaussian
* Student's t
* Johnson Su
* Normal Inverse Gaussian
* Bivariate Copula
* Gaussian Copula
* Student's t Copula
* Clayton Copula
* Gumbel Copula
* Joe Copula
* Independent Copula
* Multivariate
* Vine Copula (Regular, Centered, Clustered, Conditional Sampling)
* Prior Estimator:
* Empirical
* Black & Litterman
* Factor Model
* Synthetic Data (Stress Test, Factor Stress Test)
* Entropy Pooling
* Opinion Pooling
* Uncertainty Set Estimator:
* On Expected Returns:
* Empirical
* Circular Bootstrap
* On Covariance:
* Empirical
* Circular Bootstrap
* Pre-Selection Transformer:
* Non-Dominated Selection
* Select K Extremes (Best or Worst)
* Drop Highly Correlated Assets
* Select Non-Expiring Assets
* Select Complete Assets (handle late inception, delisting, etc.)
* Drop Zero Variance
* Cross-Validation and Model Selection:
* Compatible with all `sklearn` methods (KFold, etc.)
* Walk Forward
* Combinatorial Purged Cross-Validation
* Multiple Randomized Cross-Validation
* Hyper-Parameter Tuning:
* Compatible with all `sklearn` methods (GridSearchCV, RandomizedSearchCV)
* Risk Measures:
* Variance
* Semi-Variance
* Mean Absolute Deviation
* First Lower Partial Moment
* CVaR (Conditional Value at Risk)
* EVaR (Entropic Value at Risk)
* Worst Realization
* CDaR (Conditional Drawdown at Risk)
* Maximum Drawdown
* Average Drawdown
* EDaR (Entropic Drawdown at Risk)
* Ulcer Index
* Gini Mean Difference
* Value at Risk
* Drawdown at Risk
* Entropic Risk Measure
* Fourth Central Moment
* Fourth Lower Partial Moment
* Skew
* Kurtosis
* Optimization Features:
* Minimize Risk
* Maximize Returns
* Maximize Utility
* Maximize Ratio
* Transaction Costs
* Management Fees
* L1 and L2 Regularization
* Weight Constraints
* Group Constraints
* Budget Constraints
* Tracking Error Constraints
* Turnover Constraints
* Cardinality and Group Cardinality Constraints
* Threshold (Long and Short) Constraints
Quickstart
~~~~~~~~~~
The code snippets below are designed to introduce the functionality of `skfolio` so you
can start using it quickly. It follows the same API as scikit-learn.
Imports
-------
.. code-block:: python
from sklearn import set_config
from sklearn.model_selection import (
GridSearchCV,
KFold,
RandomizedSearchCV,
train_test_split,
)
from sklearn.pipeline import Pipeline
from scipy.stats import loguniform
from skfolio import RatioMeasure, RiskMeasure
from skfolio.datasets import load_factors_dataset, load_sp500_dataset
from skfolio.distribution import VineCopula
from skfolio.model_selection import (
CombinatorialPurgedCV,
WalkForward,
cross_val_predict,
)
from skfolio.moments import (
DenoiseCovariance,
DetoneCovariance,
EWMu,
GerberCovariance,
ShrunkMu,
)
from skfolio.optimization import (
MeanRisk,
HierarchicalRiskParity,
NestedClustersOptimization,
ObjectiveFunction,
RiskBudgeting,
)
from skfolio.pre_selection import SelectKExtremes
from skfolio.preprocessing import prices_to_returns
from skfolio.prior import (
BlackLitterman,
EmpiricalPrior,
EntropyPooling,
FactorModel,
OpinionPooling,
SyntheticData,
)
from skfolio.uncertainty_set import BootstrapMuUncertaintySet
Load Dataset
------------
.. code-block:: python
prices = load_sp500_dataset()
Train/Test split
----------------
.. code-block:: python
X = prices_to_returns(prices)
X_train, X_test = train_test_split(X, test_size=0.33, shuffle=False)
Minimum Variance
----------------
.. code-block:: python
model = MeanRisk()
Fit on Training Set
-------------------
.. code-block:: python
model.fit(X_train)
print(model.weights_)
Predict on Test Set
-------------------
.. code-block:: python
portfolio = model.predict(X_test)
print(portfolio.annualized_sharpe_ratio)
print(portfolio.summary())
Maximum Sortino Ratio
---------------------
.. code-block:: python
model = MeanRisk(
objective_function=ObjectiveFunction.MAXIMIZE_RATIO,
risk_measure=RiskMeasure.SEMI_VARIANCE,
)
Denoised Covariance & Shrunk Expected Returns
---------------------------------------------
.. code-block:: python
model = MeanRisk(
objective_function=ObjectiveFunction.MAXIMIZE_RATIO,
prior_estimator=EmpiricalPrior(
mu_estimator=ShrunkMu(), covariance_estimator=DenoiseCovariance()
),
)
Uncertainty Set on Expected Returns
-----------------------------------
.. code-block:: python
model = MeanRisk(
objective_function=ObjectiveFunction.MAXIMIZE_RATIO,
mu_uncertainty_set_estimator=BootstrapMuUncertaintySet(),
)
Weight Constraints & Transaction Costs
--------------------------------------
.. code-block:: python
model = MeanRisk(
min_weights={"AAPL": 0.10, "JPM": 0.05},
max_weights=0.8,
transaction_costs={"AAPL": 0.0001, "RRC": 0.0002},
groups=[
["Equity"] * 3 + ["Fund"] * 5 + ["Bond"] * 12,
["US"] * 2 + ["Europe"] * 8 + ["Japan"] * 10,
],
linear_constraints=[
"Equity <= 0.5 * Bond",
"US >= 0.1",
"Europe >= 0.5 * Fund",
"Japan <= 1",
],
)
model.fit(X_train)
Risk Parity on CVaR
-------------------
.. code-block:: python
model = RiskBudgeting(risk_measure=RiskMeasure.CVAR)
Risk Parity & Gerber Covariance
-------------------------------
.. code-block:: python
model = RiskBudgeting(
prior_estimator=EmpiricalPrior(covariance_estimator=GerberCovariance())
)
Nested Cluster Optimization with Cross-Validation and Parallelization
---------------------------------------------------------------------
.. code-block:: python
model = NestedClustersOptimization(
inner_estimator=MeanRisk(risk_measure=RiskMeasure.CVAR),
outer_estimator=RiskBudgeting(risk_measure=RiskMeasure.VARIANCE),
cv=KFold(),
n_jobs=-1,
)
Randomized Search of the L2 Norm
--------------------------------
.. code-block:: python
randomized_search = RandomizedSearchCV(
estimator=MeanRisk(),
cv=WalkForward(train_size=252, test_size=60),
param_distributions={
"l2_coef": loguniform(1e-3, 1e-1),
},
)
randomized_search.fit(X_train)
best_model = randomized_search.best_estimator_
print(best_model.weights_)
Grid Search on Embedded Parameters
----------------------------------
.. code-block:: python
model = MeanRisk(
objective_function=ObjectiveFunction.MAXIMIZE_RATIO,
risk_measure=RiskMeasure.VARIANCE,
prior_estimator=EmpiricalPrior(mu_estimator=EWMu(alpha=0.2)),
)
print(model.get_params(deep=True))
gs = GridSearchCV(
estimator=model,
cv=KFold(n_splits=5, shuffle=False),
n_jobs=-1,
param_grid={
"risk_measure": [
RiskMeasure.VARIANCE,
RiskMeasure.CVAR,
RiskMeasure.VARIANCE.CDAR,
],
"prior_estimator__mu_estimator__alpha": [0.05, 0.1, 0.2, 0.5],
},
)
gs.fit(X)
best_model = gs.best_estimator_
print(best_model.weights_)
Black & Litterman Model
-----------------------
.. code-block:: python
views = ["AAPL - BBY == 0.03 ", "CVX - KO == 0.04", "MSFT == 0.06 "]
model = MeanRisk(
objective_function=ObjectiveFunction.MAXIMIZE_RATIO,
prior_estimator=BlackLitterman(views=views),
)
Factor Model
------------
.. code-block:: python
factor_prices = load_factors_dataset()
X, factors = prices_to_returns(prices, factor_prices)
X_train, X_test, factors_train, factors_test = train_test_split(
X, factors, test_size=0.33, shuffle=False
)
model = MeanRisk(prior_estimator=FactorModel())
model.fit(X_train, factors_train)
print(model.weights_)
portfolio = model.predict(X_test)
print(portfolio.calmar_ratio)
print(portfolio.summary())
Factor Model & Covariance Detoning
----------------------------------
.. code-block:: python
model = MeanRisk(
prior_estimator=FactorModel(
factor_prior_estimator=EmpiricalPrior(covariance_estimator=DetoneCovariance())
)
)
Black & Litterman Factor Model
------------------------------
.. code-block:: python
factor_views = ["MTUM - QUAL == 0.03 ", "VLUE == 0.06"]
model = MeanRisk(
objective_function=ObjectiveFunction.MAXIMIZE_RATIO,
prior_estimator=FactorModel(
factor_prior_estimator=BlackLitterman(views=factor_views),
),
)
Pre-Selection Pipeline
----------------------
.. code-block:: python
set_config(transform_output="pandas")
model = Pipeline(
[
("pre_selection", SelectKExtremes(k=10, highest=True)),
("optimization", MeanRisk()),
]
)
model.fit(X_train)
portfolio = model.predict(X_test)
K-fold Cross-Validation
-----------------------
.. code-block:: python
model = MeanRisk()
mmp = cross_val_predict(model, X_test, cv=KFold(n_splits=5))
# mmp is the predicted MultiPeriodPortfolio object composed of 5 Portfolios (1 per testing fold)
mmp.plot_cumulative_returns()
print(mmp.summary())
Combinatorial Purged Cross-Validation
-------------------------------------
.. code-block:: python
model = MeanRisk()
cv = CombinatorialPurgedCV(n_folds=10, n_test_folds=2)
print(cv.summary(X_train))
population = cross_val_predict(model, X_train, cv=cv)
population.plot_distribution(
measure_list=[RatioMeasure.SHARPE_RATIO, RatioMeasure.SORTINO_RATIO]
)
population.plot_cumulative_returns()
print(population.summary())
Minimum CVaR Optimization on Synthetic Returns
----------------------------------------------
.. code-block:: python
vine = VineCopula(log_transform=True, n_jobs=-1)
prior = SyntheticData(distribution_estimator=vine, n_samples=2000)
model = MeanRisk(risk_measure=RiskMeasure.CVAR, prior_estimator=prior)
model.fit(X)
print(model.weights_)
Stress Test
-----------
.. code-block:: python
vine = VineCopula(log_transform=True, central_assets=["BAC"], n_jobs=-1)
vine.fit(X)
X_stressed = vine.sample(n_samples=10_000, conditioning = {"BAC": -0.2})
ptf_stressed = model.predict(X_stressed)
Minimum CVaR Optimization on Synthetic Factors
----------------------------------------------
.. code-block:: python
vine = VineCopula(central_assets=["QUAL"], log_transform=True, n_jobs=-1)
factor_prior = SyntheticData(
distribution_estimator=vine,
n_samples=10_000,
sample_args=dict(conditioning={"QUAL": -0.2}),
)
factor_model = FactorModel(factor_prior_estimator=factor_prior)
model = MeanRisk(risk_measure=RiskMeasure.CVAR, prior_estimator=factor_model)
model.fit(X, factors)
print(model.weights_)
Factor Stress Test
------------------
.. code-block:: python
factor_model.set_params(factor_prior_estimator__sample_args=dict(
conditioning={"QUAL": -0.5}
))
factor_model.fit(X, factors)
stressed_dist = factor_model.return_distribution_
stressed_ptf = model.predict(stressed_dist)
Entropy Pooling
---------------
.. code-block:: python
entropy_pooling = EntropyPooling(
mean_views=[
"JPM == -0.002",
"PG >= LLY",
"BAC >= prior(BAC) * 1.2",
],
cvar_views=[
"GE == 0.08",
],
)
entropy_pooling.fit(X)
print(entropy_pooling.relative_entropy_)
print(entropy_pooling.effective_number_of_scenarios_)
print(entropy_pooling.return_distribution_.sample_weight)
CVaR Hierarchical Risk Parity optimization on Entropy Pooling
-------------------------------------------------------------
.. code-block:: python
entropy_pooling = EntropyPooling(cvar_views=["GE == 0.08"])
model = HierarchicalRiskParity(
risk_measure=RiskMeasure.CVAR,
prior_estimator=entropy_pooling
)
model.fit(X)
print(model.weights_)
Stress Test with Entropy Pooling on Factor Synthetic Data
---------------------------------------------------------
.. code-block:: python
# Regular Vine Copula and sampling of 100,000 synthetic factor returns
factor_synth = SyntheticData(
n_samples=100_000,
distribution_estimator=VineCopula(log_transform=True, n_jobs=-1, random_state=0)
)
# Entropy Pooling by imposing a CVaR-95% of 10% on the Quality factor
factor_entropy_pooling = EntropyPooling(
prior_estimator=factor_synth,
cvar_views=["QUAL == 0.10"],
)
factor_entropy_pooling.fit(X, factors)
# We retrieve the stressed distribution:
stressed_dist = factor_model.return_distribution_
# We stress-test our portfolio:
stressed_ptf = model.predict(stressed_dist)
Opinion Pooling
---------------
.. code-block:: python
# We consider two expert opinions, each generated via Entropy Pooling with
# user-defined views.
# We assign probabilities of 40% to Expert 1, 50% to Expert 2, and by default
# the remaining 10% is allocated to the prior distribution:
opinion_1 = EntropyPooling(cvar_views=["AMD == 0.10"])
opinion_2 = EntropyPooling(
mean_views=["AMD >= BAC", "JPM <= prior(JPM) * 0.8"],
cvar_views=["GE == 0.12"],
)
opinion_pooling = OpinionPooling(
estimators=[("opinion_1", opinion_1), ("opinion_2", opinion_2)],
opinion_probabilities=[0.4, 0.5],
)
opinion_pooling.fit(X)
Recognition
~~~~~~~~~~~
We would like to thank all contributors to our direct dependencies, such as
`scikit-learn <https://github.com/scikit-learn/scikit-learn>`_ and `cvxpy <https://github.com/cvxpy/cvxpy>`_, as well as the contributors of the following resources that
served as sources of inspiration:
* PyPortfolioOpt
* Riskfolio-Lib
* scikit-portfolio
* microprediction
* statsmodels
* rsome
* danielppalomar.com
* gautier.marti.ai
Citation
~~~~~~~~
If you use `skfolio` in a scientific publication, we would appreciate citations:
**The library:**
.. code-block:: bibtex
@software{skfolio,
title = {skfolio},
author = {Delatte, Hugo and Nicolini, Carlo and Manzi, Matteo},
year = {2024},
doi = {10.5281/zenodo.16148630},
url = {https://doi.org/10.5281/zenodo.16148630}
}
The above uses the concept DOI, which always resolves to the latest release.
If you need precise reproducibility, especially for journals or conferences that require
it, you can cite the version-specific DOI for the exact release you used. To find it,
go to our `Zenodo project page <https://doi.org/10.5281/zenodo.16148630>`_, locate the
release you wish to reference (e.g. "v0.10.2"), and copy the DOI listed next to that
version.
**The paper:**
.. code-block:: bibtex
@article{nicolini2025skfolio,
title = {skfolio: Portfolio Optimization in Python},
author = {Nicolini, Carlo and Manzi, Matteo and Delatte, Hugo},
journal = {arXiv preprint arXiv:2507.04176},
year = {2025},
eprint = {2507.04176},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2507.04176}
}
Raw data
{
"_id": null,
"home_page": null,
"name": "skfolio",
"maintainer": null,
"docs_url": null,
"requires_python": ">=3.10",
"maintainer_email": "Hugo Delatte <delatte.hugo@gmail.com>, Matteo Manzi <matteomanzi09@gmail.com>, Carlo Nicolini <c.nicolini@ipazia.com>",
"keywords": "portfolio, optimization, optimisation, finance, asset, allocation, quantitative, quant, investment, strategy, machine-learning, scikit-learn, data-mining, data-science",
"author": null,
"author_email": "Hugo Delatte <delatte.hugo@gmail.com>",
"download_url": "https://files.pythonhosted.org/packages/08/93/69068c6d1154381f26346d8198245fd29978b2c5afab458f21076fad63a4/skfolio-0.11.0.tar.gz",
"platform": null,
"description": ".. -*- mode: rst -*-\n\n|Licence| |Codecov| |Black| |PythonVersion| |PyPi| |CI/CD| |Downloads| |Ruff| |Contribution| |Website| |JupyterLite| |Discord| |DOI|\n\n.. |Licence| image:: https://img.shields.io/badge/License-BSD%203--Clause-blue.svg\n :target: https://github.com/skfolio/skfolio/blob/main/LICENSE\n\n.. |Codecov| image:: https://codecov.io/gh/skfolio/skfolio/graph/badge.svg?token=KJ0SE4LHPV\n :target: https://codecov.io/gh/skfolio/skfolio\n\n.. |PythonVersion| image:: https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-blue.svg\n :target: https://pypi.org/project/skfolio/\n\n.. |PyPi| image:: https://img.shields.io/pypi/v/skfolio\n :target: https://pypi.org/project/skfolio\n\n.. |Black| image:: https://img.shields.io/badge/code%20style-black-000000.svg\n :target: https://github.com/psf/black\n\n.. |CI/CD| image:: https://img.shields.io/github/actions/workflow/status/skfolio/skfolio/release.yml.svg?logo=github\n :target: https://github.com/skfolio/skfolio/raw/main/LICENSE\n\n.. |Downloads| image:: https://static.pepy.tech/badge/skfolio\n :target: https://pepy.tech/project/skfolio\n\n.. |Ruff| image:: https://img.shields.io/endpoint?url=https://raw.githubusercontent.com/astral-sh/ruff/main/assets/badge/v2.json\n :target: https://github.com/astral-sh/ruff\n\n.. |Contribution| image:: https://img.shields.io/badge/Contributions-Welcome-blue\n :target: https://github.com/skfolio/skfolio/blob/main/CONTRIBUTING.md\n\n.. |Website| image:: https://img.shields.io/website.svg?down_color=red&down_message=down&up_color=53cc0d&up_message=up&url=https://skfolio.org\n :target: https://skfolio.org\n\n.. |JupyterLite| image:: https://jupyterlite.rtfd.io/en/latest/_static/badge.svg\n :target: https://skfolio.org/lite\n\n.. |Discord| image:: https://img.shields.io/badge/Discord-Join%20Chat-5865F2?logo=discord&logoColor=white\n :target: https://discord.gg/Bu7EtNYugS\n\n.. |DOI| image:: https://zenodo.org/badge/731792488.svg\n :target: https://doi.org/10.5281/zenodo.16148630\n\n.. |PythonMinVersion| replace:: 3.10\n.. |NumpyMinVersion| replace:: 1.23.4\n.. |ScipyMinVersion| replace:: 1.8.0\n.. |PandasMinVersion| replace:: 1.4.1\n.. |CvxpyBaseMinVersion| replace:: 1.5.0\n.. |ClarabelMinVersion| replace:: 0.9.0\n.. |SklearnMinVersion| replace:: 1.6.0\n.. |JoblibMinVersion| replace:: 1.3.2\n.. |PlotlyMinVersion| replace:: 5.22.0\n\n\n===============\n|icon| skfolio\n===============\n.. |icon| image:: https://raw.githubusercontent.com/skfolio/skfolio/master/docs/_static/logo_animate.svg\n :width: 100\n :alt: skfolio documentation\n :target: https://skfolio.org/\n\n\n**skfolio** is a Python library for portfolio optimization and risk management built on\ntop of scikit-learn. It offers a unified interface and tools compatible with\nscikit-learn to build, fine-tune, cross-validate and stress-test portfolio models.\n\nIt is distributed under the open-source 3-Clause BSD license.\n\n.. image:: https://raw.githubusercontent.com/skfolio/skfolio/master/docs/_static/expo.jpg\n :target: https://skfolio.org/auto_examples/\n :alt: examples\n\nImportant links\n~~~~~~~~~~~~~~~\n\n- Documentation: https://skfolio.org/\n- Examples: https://skfolio.org/auto_examples/\n- User Guide: https://skfolio.org/user_guide/\n- GitHub Repo: https://github.com/skfolio/skfolio/\n\nInstallation\n~~~~~~~~~~~~\n\n`skfolio` is available on PyPI and can be installed with::\n\n pip install -U skfolio\n\n\nDependencies\n~~~~~~~~~~~~\n\n`skfolio` requires:\n\n- python (>= |PythonMinVersion|)\n- numpy (>= |NumpyMinVersion|)\n- scipy (>= |ScipyMinVersion|)\n- pandas (>= |PandasMinVersion|)\n- cvxpy-base (>= |CvxpyBaseMinVersion|)\n- clarabel (>= |ClarabelMinVersion|)\n- scikit-learn (>= |SklearnMinVersion|)\n- joblib (>= |JoblibMinVersion|)\n- plotly (>= |PlotlyMinVersion|)\n\nDocker\n~~~~~~\n\nYou can also spin up a reproducible JupyterLab environment using Docker:\n\nBuild the image::\n\n docker build -t skfolio-jupyterlab .\n\nRun the container::\n\n docker run -p 8888:8888 -v <path-to-your-folder-containing-data>:/app/data -it skfolio-jupyterlab\n\nBrowse:\n\nOpen localhost:8888/lab and start using `skfolio`\n\nKey Concepts\n~~~~~~~~~~~~\nSince the development of modern portfolio theory by Markowitz (1952), mean-variance\noptimization (MVO) has received considerable attention.\n\nUnfortunately, it faces a number of shortcomings, including high sensitivity to the\ninput parameters (expected returns and covariance), weight concentration, high turnover,\nand poor out-of-sample performance.\n\nIt is well-known that naive allocation (1/N, inverse-vol, etc.) tends to outperform\nMVO out-of-sample (DeMiguel, 2007).\n\nNumerous approaches have been developed to alleviate these shortcomings (shrinkage,\nadditional constraints, regularization, uncertainty set, higher moments, Bayesian\napproaches, coherent risk measures, left-tail risk optimization, distributionally robust\noptimization, factor model, risk-parity, hierarchical clustering, ensemble methods,\npre-selection, etc.).\n\nGiven the large number of methods, and the fact that they can be combined, there is a\nneed for a unified framework with a machine-learning approach to perform model\nselection, validation, and parameter tuning while mitigating the risk of data leakage\nand overfitting.\n\nThis framework is built on scikit-learn's API.\n\nAvailable models\n~~~~~~~~~~~~~~~~\n\n* Portfolio Optimization:\n * Naive:\n * Equal-Weighted\n * Inverse-Volatility\n * Random (Dirichlet)\n * Convex:\n * Mean-Risk\n * Risk Budgeting\n * Maximum Diversification\n * Distributionally Robust CVaR\n * Clustering:\n * Hierarchical Risk Parity\n * Hierarchical Equal Risk Contribution\n * Nested Clusters Optimization\n * Ensemble Methods:\n * Stacking Optimization\n\n* Expected Returns Estimator:\n * Empirical\n * Exponentially Weighted\n * Equilibrium\n * Shrinkage\n\n* Covariance Estimator:\n * Empirical\n * Gerber\n * Denoising\n * Detoning\n * Exponentially Weighted\n * Ledoit-Wolf\n * Oracle Approximating Shrinkage\n * Shrunk Covariance\n * Graphical Lasso CV\n * Implied Covariance\n\n* Distance Estimator:\n * Pearson Distance\n * Kendall Distance\n * Spearman Distance\n * Covariance Distance (based on any of the above covariance estimators)\n * Distance Correlation\n * Variation of Information\n\n* Distribution Estimator:\n * Univariate:\n * Gaussian\n * Student's t\n * Johnson Su\n * Normal Inverse Gaussian\n * Bivariate Copula\n * Gaussian Copula\n * Student's t Copula\n * Clayton Copula\n * Gumbel Copula\n * Joe Copula\n * Independent Copula\n * Multivariate\n * Vine Copula (Regular, Centered, Clustered, Conditional Sampling)\n\n* Prior Estimator:\n * Empirical\n * Black & Litterman\n * Factor Model\n * Synthetic Data (Stress Test, Factor Stress Test)\n * Entropy Pooling\n * Opinion Pooling\n\n* Uncertainty Set Estimator:\n * On Expected Returns:\n * Empirical\n * Circular Bootstrap\n * On Covariance:\n * Empirical\n * Circular Bootstrap\n\n* Pre-Selection Transformer:\n * Non-Dominated Selection\n * Select K Extremes (Best or Worst)\n * Drop Highly Correlated Assets\n * Select Non-Expiring Assets\n * Select Complete Assets (handle late inception, delisting, etc.)\n * Drop Zero Variance\n\n* Cross-Validation and Model Selection:\n * Compatible with all `sklearn` methods (KFold, etc.)\n * Walk Forward\n * Combinatorial Purged Cross-Validation\n * Multiple Randomized Cross-Validation\n\n* Hyper-Parameter Tuning:\n * Compatible with all `sklearn` methods (GridSearchCV, RandomizedSearchCV)\n\n* Risk Measures:\n * Variance\n * Semi-Variance\n * Mean Absolute Deviation\n * First Lower Partial Moment\n * CVaR (Conditional Value at Risk)\n * EVaR (Entropic Value at Risk)\n * Worst Realization\n * CDaR (Conditional Drawdown at Risk)\n * Maximum Drawdown\n * Average Drawdown\n * EDaR (Entropic Drawdown at Risk)\n * Ulcer Index\n * Gini Mean Difference\n * Value at Risk\n * Drawdown at Risk\n * Entropic Risk Measure\n * Fourth Central Moment\n * Fourth Lower Partial Moment\n * Skew\n * Kurtosis\n\n* Optimization Features:\n * Minimize Risk\n * Maximize Returns\n * Maximize Utility\n * Maximize Ratio\n * Transaction Costs\n * Management Fees\n * L1 and L2 Regularization\n * Weight Constraints\n * Group Constraints\n * Budget Constraints\n * Tracking Error Constraints\n * Turnover Constraints\n * Cardinality and Group Cardinality Constraints\n * Threshold (Long and Short) Constraints\n\nQuickstart\n~~~~~~~~~~\nThe code snippets below are designed to introduce the functionality of `skfolio` so you\ncan start using it quickly. It follows the same API as scikit-learn.\n\nImports\n-------\n.. code-block:: python\n\n from sklearn import set_config\n from sklearn.model_selection import (\n GridSearchCV,\n KFold,\n RandomizedSearchCV,\n train_test_split,\n )\n from sklearn.pipeline import Pipeline\n from scipy.stats import loguniform\n\n from skfolio import RatioMeasure, RiskMeasure\n from skfolio.datasets import load_factors_dataset, load_sp500_dataset\n from skfolio.distribution import VineCopula\n from skfolio.model_selection import (\n CombinatorialPurgedCV,\n WalkForward,\n cross_val_predict,\n )\n from skfolio.moments import (\n DenoiseCovariance,\n DetoneCovariance,\n EWMu,\n GerberCovariance,\n ShrunkMu,\n )\n from skfolio.optimization import (\n MeanRisk,\n HierarchicalRiskParity,\n NestedClustersOptimization,\n ObjectiveFunction,\n RiskBudgeting,\n )\n from skfolio.pre_selection import SelectKExtremes\n from skfolio.preprocessing import prices_to_returns\n from skfolio.prior import (\n BlackLitterman,\n EmpiricalPrior,\n EntropyPooling,\n FactorModel,\n OpinionPooling,\n SyntheticData,\n )\n from skfolio.uncertainty_set import BootstrapMuUncertaintySet\n\nLoad Dataset\n------------\n.. code-block:: python\n\n prices = load_sp500_dataset()\n\nTrain/Test split\n----------------\n.. code-block:: python\n\n X = prices_to_returns(prices)\n X_train, X_test = train_test_split(X, test_size=0.33, shuffle=False)\n\n\nMinimum Variance\n----------------\n.. code-block:: python\n\n model = MeanRisk()\n\nFit on Training Set\n-------------------\n.. code-block:: python\n\n model.fit(X_train)\n\n print(model.weights_)\n\nPredict on Test Set\n-------------------\n.. code-block:: python\n\n portfolio = model.predict(X_test)\n\n print(portfolio.annualized_sharpe_ratio)\n print(portfolio.summary())\n\n\n\nMaximum Sortino Ratio\n---------------------\n.. code-block:: python\n\n model = MeanRisk(\n objective_function=ObjectiveFunction.MAXIMIZE_RATIO,\n risk_measure=RiskMeasure.SEMI_VARIANCE,\n )\n\n\nDenoised Covariance & Shrunk Expected Returns\n---------------------------------------------\n.. code-block:: python\n\n model = MeanRisk(\n objective_function=ObjectiveFunction.MAXIMIZE_RATIO,\n prior_estimator=EmpiricalPrior(\n mu_estimator=ShrunkMu(), covariance_estimator=DenoiseCovariance()\n ),\n )\n\nUncertainty Set on Expected Returns\n-----------------------------------\n.. code-block:: python\n\n model = MeanRisk(\n objective_function=ObjectiveFunction.MAXIMIZE_RATIO,\n mu_uncertainty_set_estimator=BootstrapMuUncertaintySet(),\n )\n\n\nWeight Constraints & Transaction Costs\n--------------------------------------\n.. code-block:: python\n\n model = MeanRisk(\n min_weights={\"AAPL\": 0.10, \"JPM\": 0.05},\n max_weights=0.8,\n transaction_costs={\"AAPL\": 0.0001, \"RRC\": 0.0002},\n groups=[\n [\"Equity\"] * 3 + [\"Fund\"] * 5 + [\"Bond\"] * 12,\n [\"US\"] * 2 + [\"Europe\"] * 8 + [\"Japan\"] * 10,\n ],\n linear_constraints=[\n \"Equity <= 0.5 * Bond\",\n \"US >= 0.1\",\n \"Europe >= 0.5 * Fund\",\n \"Japan <= 1\",\n ],\n )\n model.fit(X_train)\n\n\nRisk Parity on CVaR\n-------------------\n.. code-block:: python\n\n model = RiskBudgeting(risk_measure=RiskMeasure.CVAR)\n\nRisk Parity & Gerber Covariance\n-------------------------------\n.. code-block:: python\n\n model = RiskBudgeting(\n prior_estimator=EmpiricalPrior(covariance_estimator=GerberCovariance())\n )\n\nNested Cluster Optimization with Cross-Validation and Parallelization\n---------------------------------------------------------------------\n.. code-block:: python\n\n model = NestedClustersOptimization(\n inner_estimator=MeanRisk(risk_measure=RiskMeasure.CVAR),\n outer_estimator=RiskBudgeting(risk_measure=RiskMeasure.VARIANCE),\n cv=KFold(),\n n_jobs=-1,\n )\n\nRandomized Search of the L2 Norm\n--------------------------------\n.. code-block:: python\n\n randomized_search = RandomizedSearchCV(\n estimator=MeanRisk(),\n cv=WalkForward(train_size=252, test_size=60),\n param_distributions={\n \"l2_coef\": loguniform(1e-3, 1e-1),\n },\n )\n randomized_search.fit(X_train)\n\n best_model = randomized_search.best_estimator_\n\n print(best_model.weights_)\n\n\nGrid Search on Embedded Parameters\n----------------------------------\n.. code-block:: python\n\n model = MeanRisk(\n objective_function=ObjectiveFunction.MAXIMIZE_RATIO,\n risk_measure=RiskMeasure.VARIANCE,\n prior_estimator=EmpiricalPrior(mu_estimator=EWMu(alpha=0.2)),\n )\n\n print(model.get_params(deep=True))\n\n gs = GridSearchCV(\n estimator=model,\n cv=KFold(n_splits=5, shuffle=False),\n n_jobs=-1,\n param_grid={\n \"risk_measure\": [\n RiskMeasure.VARIANCE,\n RiskMeasure.CVAR,\n RiskMeasure.VARIANCE.CDAR,\n ],\n \"prior_estimator__mu_estimator__alpha\": [0.05, 0.1, 0.2, 0.5],\n },\n )\n gs.fit(X)\n\n best_model = gs.best_estimator_\n\n print(best_model.weights_)\n\n\nBlack & Litterman Model\n-----------------------\n.. code-block:: python\n\n views = [\"AAPL - BBY == 0.03 \", \"CVX - KO == 0.04\", \"MSFT == 0.06 \"]\n model = MeanRisk(\n objective_function=ObjectiveFunction.MAXIMIZE_RATIO,\n prior_estimator=BlackLitterman(views=views),\n )\n\nFactor Model\n------------\n.. code-block:: python\n\n factor_prices = load_factors_dataset()\n\n X, factors = prices_to_returns(prices, factor_prices)\n X_train, X_test, factors_train, factors_test = train_test_split(\n X, factors, test_size=0.33, shuffle=False\n )\n\n model = MeanRisk(prior_estimator=FactorModel())\n model.fit(X_train, factors_train)\n\n print(model.weights_)\n\n portfolio = model.predict(X_test)\n\n print(portfolio.calmar_ratio)\n print(portfolio.summary())\n\nFactor Model & Covariance Detoning\n----------------------------------\n.. code-block:: python\n\n model = MeanRisk(\n prior_estimator=FactorModel(\n factor_prior_estimator=EmpiricalPrior(covariance_estimator=DetoneCovariance())\n )\n )\n\nBlack & Litterman Factor Model\n------------------------------\n.. code-block:: python\n\n factor_views = [\"MTUM - QUAL == 0.03 \", \"VLUE == 0.06\"]\n model = MeanRisk(\n objective_function=ObjectiveFunction.MAXIMIZE_RATIO,\n prior_estimator=FactorModel(\n factor_prior_estimator=BlackLitterman(views=factor_views),\n ),\n )\n\nPre-Selection Pipeline\n----------------------\n.. code-block:: python\n\n set_config(transform_output=\"pandas\")\n model = Pipeline(\n [\n (\"pre_selection\", SelectKExtremes(k=10, highest=True)),\n (\"optimization\", MeanRisk()),\n ]\n )\n model.fit(X_train)\n\n portfolio = model.predict(X_test)\n\n\n\n\nK-fold Cross-Validation\n-----------------------\n.. code-block:: python\n\n model = MeanRisk()\n mmp = cross_val_predict(model, X_test, cv=KFold(n_splits=5))\n # mmp is the predicted MultiPeriodPortfolio object composed of 5 Portfolios (1 per testing fold)\n\n mmp.plot_cumulative_returns()\n print(mmp.summary())\n\n\nCombinatorial Purged Cross-Validation\n-------------------------------------\n.. code-block:: python\n\n model = MeanRisk()\n\n cv = CombinatorialPurgedCV(n_folds=10, n_test_folds=2)\n\n print(cv.summary(X_train))\n\n population = cross_val_predict(model, X_train, cv=cv)\n\n population.plot_distribution(\n measure_list=[RatioMeasure.SHARPE_RATIO, RatioMeasure.SORTINO_RATIO]\n )\n population.plot_cumulative_returns()\n print(population.summary())\n\n\nMinimum CVaR Optimization on Synthetic Returns\n----------------------------------------------\n.. code-block:: python\n\n vine = VineCopula(log_transform=True, n_jobs=-1)\n prior = SyntheticData(distribution_estimator=vine, n_samples=2000)\n model = MeanRisk(risk_measure=RiskMeasure.CVAR, prior_estimator=prior)\n model.fit(X)\n print(model.weights_)\n\n\nStress Test\n-----------\n.. code-block:: python\n\n vine = VineCopula(log_transform=True, central_assets=[\"BAC\"], n_jobs=-1)\n vine.fit(X)\n X_stressed = vine.sample(n_samples=10_000, conditioning = {\"BAC\": -0.2})\n ptf_stressed = model.predict(X_stressed)\n\n\nMinimum CVaR Optimization on Synthetic Factors\n----------------------------------------------\n.. code-block:: python\n\n vine = VineCopula(central_assets=[\"QUAL\"], log_transform=True, n_jobs=-1)\n factor_prior = SyntheticData(\n distribution_estimator=vine,\n n_samples=10_000,\n sample_args=dict(conditioning={\"QUAL\": -0.2}),\n )\n factor_model = FactorModel(factor_prior_estimator=factor_prior)\n model = MeanRisk(risk_measure=RiskMeasure.CVAR, prior_estimator=factor_model)\n model.fit(X, factors)\n print(model.weights_)\n\n\nFactor Stress Test\n------------------\n.. code-block:: python\n\n factor_model.set_params(factor_prior_estimator__sample_args=dict(\n conditioning={\"QUAL\": -0.5}\n ))\n factor_model.fit(X, factors)\n stressed_dist = factor_model.return_distribution_\n stressed_ptf = model.predict(stressed_dist)\n\nEntropy Pooling\n---------------\n.. code-block:: python\n\n entropy_pooling = EntropyPooling(\n mean_views=[\n \"JPM == -0.002\",\n \"PG >= LLY\",\n \"BAC >= prior(BAC) * 1.2\",\n ],\n cvar_views=[\n \"GE == 0.08\",\n ],\n )\n entropy_pooling.fit(X)\n print(entropy_pooling.relative_entropy_)\n print(entropy_pooling.effective_number_of_scenarios_)\n print(entropy_pooling.return_distribution_.sample_weight)\n\nCVaR Hierarchical Risk Parity optimization on Entropy Pooling\n-------------------------------------------------------------\n.. code-block:: python\n\n entropy_pooling = EntropyPooling(cvar_views=[\"GE == 0.08\"])\n model = HierarchicalRiskParity(\n risk_measure=RiskMeasure.CVAR,\n prior_estimator=entropy_pooling\n )\n model.fit(X)\n print(model.weights_)\n\nStress Test with Entropy Pooling on Factor Synthetic Data\n---------------------------------------------------------\n.. code-block:: python\n\n # Regular Vine Copula and sampling of 100,000 synthetic factor returns\n factor_synth = SyntheticData(\n n_samples=100_000,\n distribution_estimator=VineCopula(log_transform=True, n_jobs=-1, random_state=0)\n )\n\n # Entropy Pooling by imposing a CVaR-95% of 10% on the Quality factor\n factor_entropy_pooling = EntropyPooling(\n prior_estimator=factor_synth,\n cvar_views=[\"QUAL == 0.10\"],\n )\n\n factor_entropy_pooling.fit(X, factors)\n\n # We retrieve the stressed distribution:\n stressed_dist = factor_model.return_distribution_\n\n # We stress-test our portfolio:\n stressed_ptf = model.predict(stressed_dist)\n\nOpinion Pooling\n---------------\n.. code-block:: python\n\n # We consider two expert opinions, each generated via Entropy Pooling with\n # user-defined views.\n # We assign probabilities of 40% to Expert 1, 50% to Expert 2, and by default\n # the remaining 10% is allocated to the prior distribution:\n opinion_1 = EntropyPooling(cvar_views=[\"AMD == 0.10\"])\n opinion_2 = EntropyPooling(\n mean_views=[\"AMD >= BAC\", \"JPM <= prior(JPM) * 0.8\"],\n cvar_views=[\"GE == 0.12\"],\n )\n\n opinion_pooling = OpinionPooling(\n estimators=[(\"opinion_1\", opinion_1), (\"opinion_2\", opinion_2)],\n opinion_probabilities=[0.4, 0.5],\n )\n\n opinion_pooling.fit(X)\n\n\nRecognition\n~~~~~~~~~~~\n\nWe would like to thank all contributors to our direct dependencies, such as\n`scikit-learn <https://github.com/scikit-learn/scikit-learn>`_ and `cvxpy <https://github.com/cvxpy/cvxpy>`_, as well as the contributors of the following resources that\nserved as sources of inspiration:\n\n * PyPortfolioOpt\n * Riskfolio-Lib\n * scikit-portfolio\n * microprediction\n * statsmodels\n * rsome\n * danielppalomar.com\n * gautier.marti.ai\n\n\nCitation\n~~~~~~~~\n\nIf you use `skfolio` in a scientific publication, we would appreciate citations:\n\n**The library:**\n\n.. code-block:: bibtex\n\n @software{skfolio,\n title = {skfolio},\n author = {Delatte, Hugo and Nicolini, Carlo and Manzi, Matteo},\n year = {2024},\n doi = {10.5281/zenodo.16148630},\n url = {https://doi.org/10.5281/zenodo.16148630}\n }\n\nThe above uses the concept DOI, which always resolves to the latest release.\nIf you need precise reproducibility, especially for journals or conferences that require\nit, you can cite the version-specific DOI for the exact release you used. To find it,\ngo to our `Zenodo project page <https://doi.org/10.5281/zenodo.16148630>`_, locate the\nrelease you wish to reference (e.g. \"v0.10.2\"), and copy the DOI listed next to that\nversion.\n\n**The paper:**\n\n.. code-block:: bibtex\n\n @article{nicolini2025skfolio,\n title = {skfolio: Portfolio Optimization in Python},\n author = {Nicolini, Carlo and Manzi, Matteo and Delatte, Hugo},\n journal = {arXiv preprint arXiv:2507.04176},\n year = {2025},\n eprint = {2507.04176},\n archivePrefix = {arXiv},\n url = {https://arxiv.org/abs/2507.04176}\n }\n\n",
"bugtrack_url": null,
"license": "BSD 3-Clause License\n \n Copyright (c) 2007-2023 The skfolio developers.\n All rights reserved.\n \n Redistribution and use in source and binary forms, with or without\n modification, are permitted provided that the following conditions are met:\n \n * Redistributions of source code must retain the above copyright notice, this\n list of conditions and the following disclaimer.\n \n * Redistributions in binary form must reproduce the above copyright notice,\n this list of conditions and the following disclaimer in the documentation\n and/or other materials provided with the distribution.\n \n * Neither the name of the copyright holder nor the names of its\n contributors may be used to endorse or promote products derived from\n this software without specific prior written permission.\n \n THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS \"AS IS\"\n AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE\n IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE ARE\n DISCLAIMED. IN NO EVENT SHALL THE COPYRIGHT HOLDER OR CONTRIBUTORS BE LIABLE\n FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL\n DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR\n SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER\n CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,\n OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE\n OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE.",
"summary": "Portfolio optimization built on top of scikit-learn",
"version": "0.11.0",
"project_urls": {
"API Reference": "https://www.skfolio.org/api_reference.html",
"Documentation": "https://www.skfolio.org",
"Repository": "https://github.com/skfolio/skfolio",
"Tutorials": "https://www.skfolio.org"
},
"split_keywords": [
"portfolio",
" optimization",
" optimisation",
" finance",
" asset",
" allocation",
" quantitative",
" quant",
" investment",
" strategy",
" machine-learning",
" scikit-learn",
" data-mining",
" data-science"
],
"urls": [
{
"comment_text": null,
"digests": {
"blake2b_256": "59784d72902b019a629241cc22166f033f83238c2903d7012e11c066d9fb6e13",
"md5": "5f0856f1011700983c5824fdb051a634",
"sha256": "a197be8692eb5a06d1acf3abd348d3ec458c6571758cfff03b14cacd8c36401b"
},
"downloads": -1,
"filename": "skfolio-0.11.0-py3-none-any.whl",
"has_sig": false,
"md5_digest": "5f0856f1011700983c5824fdb051a634",
"packagetype": "bdist_wheel",
"python_version": "py3",
"requires_python": ">=3.10",
"size": 849982,
"upload_time": "2025-07-26T20:36:27",
"upload_time_iso_8601": "2025-07-26T20:36:27.669192Z",
"url": "https://files.pythonhosted.org/packages/59/78/4d72902b019a629241cc22166f033f83238c2903d7012e11c066d9fb6e13/skfolio-0.11.0-py3-none-any.whl",
"yanked": false,
"yanked_reason": null
},
{
"comment_text": null,
"digests": {
"blake2b_256": "089369068c6d1154381f26346d8198245fd29978b2c5afab458f21076fad63a4",
"md5": "82f6ddab792d09d20a41069abf9df7e8",
"sha256": "e8d499550010bcc0692ad0f0b360c70ca5c180cbcd6acdaa0c0fcec35b1b3682"
},
"downloads": -1,
"filename": "skfolio-0.11.0.tar.gz",
"has_sig": false,
"md5_digest": "82f6ddab792d09d20a41069abf9df7e8",
"packagetype": "sdist",
"python_version": "source",
"requires_python": ">=3.10",
"size": 752905,
"upload_time": "2025-07-26T20:36:29",
"upload_time_iso_8601": "2025-07-26T20:36:29.785857Z",
"url": "https://files.pythonhosted.org/packages/08/93/69068c6d1154381f26346d8198245fd29978b2c5afab458f21076fad63a4/skfolio-0.11.0.tar.gz",
"yanked": false,
"yanked_reason": null
}
],
"upload_time": "2025-07-26 20:36:29",
"github": true,
"gitlab": false,
"bitbucket": false,
"codeberg": false,
"github_user": "skfolio",
"github_project": "skfolio",
"travis_ci": false,
"coveralls": false,
"github_actions": true,
"lcname": "skfolio"
}
