Cvxportfolio

cvxportfolio is a python library for portfolio optimization and simulation based on the book Multi-Period Trading via Convex Optimization (also available in print).

The documentation of the package is kindly hosted by Read the Docs at www.cvxportfolio.com.

Installation

All our source code and releases are kindly hosted by the Python Package Index. You can install the latest one with

pip install -U cvxportfolio

Testing locally

We ship our unit test suite with the pip package. After installing you can test in you local environment by

python -m unittest discover cvxportfolio

Example

In the following example market data is downloaded by a public source (Yahoo finance) and the forecasts are computed iteratively, at each point in the backtest, from past data. That is, at each point in the backtest, the policy object only operates on past data, and thus the result you get is a realistic simulation of what the strategy would have performed in the market. The simulator by default includes holding and transaction costs, using the models described in the book, and default parameters that are typical for the US stock market. The logic used matches what is described in Chapter 7 of the book. For example, returns are forecasted as the historical mean returns and covariances as historical covariances (both ignoring np.nan's). The logic used is detailed in the forecast module. Many optimizations are applied to make sure the system works well with real data.

import cvxportfolio as cvx
import matplotlib.pyplot as plt

gamma = 3       # risk aversion parameter (Chapter 4.2)
kappa = 0.05    # covariance forecast error risk parameter (Chapter 4.3)
objective = cvx.ReturnsForecast() - gamma * (
	cvx.FullCovariance() + kappa * cvx.RiskForecastError()
) - cvx.StocksTransactionCost()
constraints = [cvx.LeverageLimit(3)]

policy = cvx.MultiPeriodOptimization(objective, constraints, planning_horizon=2)

simulator = cvx.StockMarketSimulator(['AAPL', 'AMZN', 'TSLA', 'GM', 'CVX', 'NKE'])

result = simulator.backtest(policy, start_time='2020-01-01')

# print backtest result statistics
print(result)

# plot backtest results
result.plot()

Development

To set up a development environment locally you should

git clone https://github.com/cvxgrp/cvxportfolio.git
cd cvxportfolio
make env

This will replicate our development environment. From there you can test with

make test

You activate the shell environment with one of scripts in env/bin (or env\Scripts on windows), for example if you use bash on POSIX

source env/bin/activate

and from the environment you can run any of the scripts in the examples (the cvxportfolio package is installed in editable mode). Or, if you don't want to activate the environment, you can just run scripts directly using env/bin/python or env\Scripts\python on windows, like we do in the Makefile.

Examples from the book

In branch 0.0.X you can find the original material used to generate plots and results in the book. As you may see from those ipython notebooks a lot of the logic that was implemented there, outside of cvxportfolio proper, is being included and made automatic in newer versions of cvxportfolio.

Academic

If you use cvxportfolio in your academic work please cite our book:

@book{BBDKKNS:17,
    author       = {S. Boyd and E. Busseti and S. Diamond and R. Kahn and K. Koh and P. Nystrup and J. Speth},
    title        = {Multi-Period Trading via Convex Optimization},
    series       = {Foundations and Trends in Optimization},
    year         = {2017},
    month        = {August},
    publisher    = {Now Publishers},
    url          = {http://stanford.edu/~boyd/papers/cvx_portfolio.html},
}

License

Cvxportfolio is licensed under the Apache 2.0 permissive open source license.