Algorithmic trading system

A system that fundamentally depends upon computerized algorithms, and the data and technological infrastructure through which they operate, to address various decisions and tasks associated with trading financial instruments.

Source: AIEOG AI Lexicon (Feb 2026), SEC Market Structure Research, HFT Literature Review (2014)

An algorithmic trading system uses computer programs to make trading decisions and execute trades automatically. These systems analyze market data, identify trading opportunities, determine order size and timing, and execute trades at speeds and volumes that would be impossible for human traders.

Algorithmic trading systems range from relatively simple rule-based programs (“buy when the 50-day moving average crosses above the 200-day moving average”) to sophisticated AI-driven systems that use machine learning to identify patterns in market data and adapt their strategies in real time.

These systems are prevalent in equities, fixed income, foreign exchange, and derivatives markets. They account for a significant share of trading volume in U.S. equity markets. As AI capabilities advance, the complexity and autonomy of these systems continue to grow.

Algorithmic trading systems are among the most regulated applications of algorithms in financial services. The SEC, FINRA, and CFTC each have rules and guidance that apply to algorithmic and high-frequency trading. Regulatory concerns center on:

• Market stability. Algorithmic trading can amplify market volatility and contribute to flash crashes. Regulators require risk controls (kill switches, circuit breakers, position limits) to prevent individual systems from destabilizing markets.
• Market manipulation. Algorithms can be used for manipulative practices like spoofing (placing orders with the intent to cancel), layering, and front-running. Regulators actively monitor for these behaviors.
• Supervision requirements. Firms that use algorithmic trading systems are expected to maintain robust supervisory controls, including pre-trade risk checks, real-time monitoring, and post-trade surveillance.
• Model risk. Trading algorithms are models under SR 11-7 and OCC guidance. They require the same governance, validation, and monitoring as any other model used for material business decisions.

1. Classify trading algorithms as models. Apply model risk management requirements to all algorithmic trading systems, including documentation, validation, and ongoing monitoring.
2. Implement pre-trade risk controls. Establish automated checks that prevent algorithms from executing trades that exceed defined risk parameters.
3. Monitor in real time. Deploy real-time surveillance that can detect anomalous trading behavior and halt systems that malfunction.
4. Maintain kill switches. Every algorithmic trading system should have a tested mechanism for immediate shutdown.
5. Document strategy logic. Maintain clear documentation of each algorithm’s trading strategy, parameters, and decision logic.
6. Test before deployment. Validate algorithms in simulated environments before live deployment, and revalidate when market conditions change significantly.

Algorithm, AI system, Anomaly detection system, Model risk, Performance monitoring, AI risk assessment