What Is High-Frequency Trading?

High-frequency trading (HFT) is a branch of algorithmic trading that focuses on generating profit using high execution speed. It’s used in areas such as arbitrage trading, signal-based trading, and scalping. In major exchanges, the trading volume generated from these trades—typically by proprietary traders, hedge fund managers, and market makers—is significant.

Developing HFT strategies requires intraday tick data and a solid analytical tool. MATLAB^® provides both. It supports popular techniques for efficiently developing, backtesting, and implementing these strategies:

Hypothesis testing, machine learning, and pattern recognition
Trading cost analysis and market impact modeling
Analysis of financial time series to generate trading signals
Monte Carlo simulation and model validation
High-performance parallel computing using GPUs, clusters, grids, and clouds

For more on tools for HFT, see MATLAB and Datafeed Toolbox™.

Examples and How To

Algorithmic Trading Strategies for Optimizing Trade Execution (58:41) - Video
Backtesting Strategy Framework in Financial Toolbox (2:17) - Video
CalPERS Analyzes Currency Market Dynamics to Identify Intraday Trading Opportunities - Customer Story
Evaluating Systematic Trading Strategies: Using MATLAB to Accelerate the Pace of Quantitative Research (32:50) - Video
Alpha Generation Using Refinitiv News Sentiment and MATLAB (59:53) - Video
Quantitative Finance and Risk Management - Overview
High-Frequency Trading Tools and Resources - MATLAB Central - MATLAB Central

Software Reference

Transaction Cost Analysis - Function
timeseries - Downloading intraday tick cata from Bloomberg - Function
Cointegration Testing - Function
Neural Network Time Series Tool - Documentation

Kissell Research Group Develops High-Frequency Trading Cost Index

Machine Learning and Big Data in Quantitative Investing

Read white paper