Computational Modelling Group

Automated Algorithmic Trading with Intelligent Execution

Started
1st September 2012
Ended
28th December 2012
Research Team
Ash Booth
Investigators
Frank McGroarty, Enrico Gerding

Results from a single experimental run showing the cumulative abnormal return (CAR) of our Adaptive Execution (AE) strategy compared to benchmarks.

Our trading agent uses techniques based on extensions of Adaptive Aggressiveness and Creamer and Freund’s boosting prediction algorithms. We find that, on top of addressing the unrealistic assumptions of previous strategies, our algorithm, on average, accounted for a 46% increase in profitability compared to the state of the art.

Categories

Algorithms and computational methods: Agents, Classification, Machine learning, statistical analysis

Visualisation and data handling software: Pylab

Software Engineering Tools: Eclipse

Programming languages and libraries: C++, Python, R

Transdisciplinary tags: Complex Systems, Computer Science, Economics