build, test, and implement statistical arbitrage trading strategies with matlab
statistical arbitrage, also referred to as stat arb, is a computationally intensive approach to algorithmically trading financial market assets such as equities and commodities. it involves the simultaneous buying and selling of security portfolios according to predefined or adaptive statistical models.
statistical arbitrage techniques are modern variations of the classic cointegration-based pairs trading strategy. this strategy is based on short-term mean reversion principles coupled with hedging strategies that take care of overall market risk.
hedge funds, mutual funds, and proprietary trading firms build, test, and implement trading strategies based on statistical arbitrage. an effective workflow entails:
- gathering data from databases and industry-standard datafeeds
- designing, testing, and optimizing trading strategies
- applying advanced statistical techniques such as machine learning
- performing cvar portfolio optimization
- connecting to trading platforms and managing an order workflow
for more information, see matlab® and toolboxes for datafeed, finance, econometrics, statistics, and optimization.
examples and how to
- machine learning for algorithmic trading (32:55) - video
- - video
- - video
- - video
- - book
- - video
software reference
- algorithmic trading code and other resources - matlab central
- - documentation
- - function
- workflow for trading technologies x_trader® - documentation
- - functions
see also: cointegration, equity trading, commodities trading, financial risk management, portfolio optimization, financial toolbox, econometrics toolbox, datafeed toolbox, swing trading