Autonomous Trade Agents

Algorithmic trading and robo-traders are on the march, but few vendors offer complete concepts and technologies beyond signalling. Algorithms usually focus on statistical arbitrage and squeezing prices rather than on statistically long-term profitable trading strategies.

Autonomy and closed-loop control for the trading industry

VERMONT HARDING's trade agents usually incorporate directional strategies. Trade agents are fully event-driven; self-adaptively, they execute their dealing actions in continuously changing and dynamic markets. Common to all our trade agents, which pursue diverse trading strategies, is to fulfil a set of statistical tests to prove their long-term profitability over time.

According to the concept of closed-loop control, each trade agent has a perception of itself, its environment and the impact of its decision-making and considers this information with each dealing decision it makes. Each trade agent can integrate with Alternative Source Fusion ASF to mediate qualitative information as well.

Deal Identification – VERMONT HARDING’s trade agents feature a proprietary multi-aspect fusion component for each asset which iteratively analyses current market moves to determine beneficial entries. Trade agents merge information of different granularity and from different angles into a holistic market picture by incorporating non-linear, non-Gaussian models to identify deal opportunities with high accuracy.

Deal Monitoring, Account and Position Management – Each trade agent monitors its positions and applies low level risk management features, such as stop losses.

Exit Management – As each trade agent is event-driven, stop loss and limit orders are being dynamically calculated and adapt dynamically when market dynamics change during runtime.

Systematic Adaptability – All models operated by VERMONT HARDING’s trade agents adapt to changing market dynamics. Rather than being optimized, they are optimizing during and modify their behavior to map with changing markets. To re-calibrate, VERMONT HARDING offers a structured re-calibration process following mathematical convergence criteria to ensure that the trade agents’ statistical footprints remain intact.

Agent Governance – Each trade agent is being governed through an Agent Management System and a networking component that allows the user graphically – by drag-and-drop – defining and managing its network of trade agents.

Logging – Each trade agent logs its functional behavior, exceptions and account movements at any given event. For major events and exceptions, a trade agent may actively notify the user through email or mobile message for manual interaction and remedy.

The Benefits of Trade Agents

The benefits of our trade agents include:

  • Statistically proven performance on the long run.
  • Each dealing opportunity is being considered and evaluated.
  • 24/7 real-time market monitoring and trading.
  • Very high degree of automation upvalues human traders to more strategic tasks.
  • Reduced operational costs for large financial service companies.

Trader Clones

Reliability and availability of their traders are critical to banking institutions. Trader clones are therefore taking on a special position with VERMONT HARDING’s customers. They are able to learn the specific behaviour of human traders and generalize it for future deals.

Contact for details: research@vermontharding.com

Next Steps