An Agent-Based Financial Market Simulator for Evaluation of Algorithmic Trading Strategies

Algorithmic trading strategies are most often evaluated by running against historical data and observing the results. This limits the evaluation scenarios to situations similar to those for which historical data is available. In order to evaluate high frequency trading systems in a broader setting, a different approach is required. This paper presents an agent-based financial market simulator that allows the exploration of market behaviour under a wide range of conditions. Agents may simulate human and algorithmic traders operating with different objectives, strategies and reaction times and market behaviour can use combinations of simulated and historical data. The simulator models the market’s structure, allowing behaviours to be specified for market makers and liquidity providers and other market participants. The primary use of the system has been in the evaluation of algorithmic trading strategies in a corporate setting, but other uses include education and training as well as policy evaluation. Keywords–Agent-Based Simulation, Financial Markets, High Frequency Trading