Artificial Intelligence in Finance

Chapter 12. Execution and Deployment

Considerable progress is needed before autonomous vehicles can operate reliably in mixed urban traffic, heavy rain and snow, unpaved and unmapped roads, and where wireless access is unreliable.

Todd Litman (2020)

An investment firm that engages in algorithmic trading shall have in place effective systems and risk controls suitable to the business it operates to ensure that its trading systems are resilient and have sufficient capacity, are subject to appropriate trading thresholds and limits and prevent the sending of erroneous orders or the systems otherwise functioning in a way that may create or contribute to a disorderly market.

MiFID II (Article 17)

Chapter 11 trains a trading bot in the form of a financial Q-learning agent based on historical data. It introduces event-based backtesting as an approach flexible enough to account for typical risk measures, such as trailing stop loss orders or take profit targets. However, all this happens asynchronously in a sandbox environment based on historical data only. As with an autonomous vehicle (AV), there is the problem of deploying the AI in the real world. For an AV this means combining the AI with the car hardware and deploying the AV on test and public streets. For a trading bot this means connecting the trading bot with a trading platform and deploying it such that orders are executed automatically. In other words, the algorithmic side is clear—execution and deployment now need to be ...