Algorithmic Trading

Develop trading systems with MATLAB

Algorithmic trading uses algorithms to drive trading decisions, usually in electronic financial markets. Applied in buy-side and sell-side institutions, algorithmic trading forms the basis of high-frequency trading, FOREX trading, and associated risk and execution analytics.

Algorithmic traders worldwide use MATLAB and add-on toolboxes to develop, backtest, and deploy mathematical models that detect and exploit market movements. With MATLAB you can perform tasks including:

• Develop trading strategies, using technical time-series, machine learning, and non-linear time-series methods
• Apply parallel and GPU computing for time-efficient back-testing and parameter identification
• Calculate profit and loss and conduct risk analysis
• Perform execution analytics, such as market impact modeling and iceberg detection
• Incorporate strategies and analytics into production trading environments

Examples and How To

• Cointegration and Pairs Trading with Econometrics Toolbox (Webinar)
• Integrating MATLAB Based Analytics Into Databases, Web, and Messaging Systems (Webinar)
• Liquidnet Develops Execution Performance Measurement Tool (User Story)
• CalPERS Analyzes Currency Market Dynamics to Identify Intraday Trading Opportunities (User Story)
• Quantitative Trading: How to Build Your Own Algorithmic Trading Business, by Ernest Chan (Book)
• Algorithmic Trading (Solution)
• Algorithmic Trading Code and Other Resources (MATLAB Central)

Software Reference

• Movavg – Leading and lagging moving averages chart (Financial Toolbox Function)
• Sharpe – compute Sharpe ratio (Financial Toolbox Function)
• Gaoptimset – Create genetic algorithm options structure (Global Optimization Toolbox Function)
• Cointegration Testing (Econometrics Toolbox Functions)
• Neural Network Time Series Tool (Neural Network Toolbox Documentation)

See also: Financial Toolbox, Econometrics Toolbox, Parallel Computing Toolbox, Global Optimization Toolbox, Neural Network Toolbox, cointegration