Skip to main content

Genetic Investment

The financial markets are subject to large and unpredictable changes. Computer algorithms that interact with financial markets have three choices: (i) adapt rapidly to change, (ii) be robust to change, or (iii) fail. But how should a Genetic Computation (GC) stock-picking system respond to volatile markets? After a market shift, the GC will need to re-train on new data, but initially there is insufficient data for re-training. Clack's research team has worked to evolve solutions that are robust to market changes:

  • Behavioural GP Diversity for Dynamic Environments: an application in hedge fund investment (Yan & Clack, in Proceedings GECCO 2006, pp 1817-1824, ACM, 2006)
  • Evolving robust GP solutions for hedge fund stock selection in emerging markets (Yan & Clack, in Proceedings GECCO 2007, pp 2234-2241,  ACM, 2007) - (Best Paper award).
  • Diverse committees vote for dependable profits (Yan & Clack, in Proceedings GECCO 2007, pp 2226-2233, ACM, 2007)
  • ALPS evaluation in Financial Portfolio Optimisation (Patel & Clack, in Proceedings IEEE CEC 2007, pp 813-819, IEEE, 2007)
  • Learning to optimize profits beats predicting returns - comparing techniques for financial portfolio optimisation (Yan, Sewell & Clack, in Proceedings GECCO 2008, pp 1681-1688, ACM Press, 2008)
  • GP Age-layer and Crossover Effects in Bid-Offer Spread Prediction (Willis, Patel & Clack, in Proceedings GECCO 2008, pp 1665-1672, ACM Press, 2008)
  • Robustness of multiple objective GP stock-picking in unstable financial markets: real-world applications track (Hassan & Clack, in Proceedings GECCO 2009, pp 1513-1520, ACM, 2009)
  • Behavioural GP diversity for adaptive stock selection (Yan & Clack, in Proceedings GECCO 2009, pp 1641-1648, ACM, 2009)

The following paper was nominated for a Best Paper award; is now cited in Hitoshi Iba's key new book surveying the field of Evolutionary Computation in Financial Engineering; was supported by Thomson Reuters; prompted seven invited talks at industry conferences; and the work was also presented at two invited talks at subsequent GECCO conferences. This novel work is the first to identify a significant instability whereby cautious investment strategies can unwittingly become adventurous, and vice versa, following a shift in market regime.

  • Multiobjective Robustness for Portfolio Optimization in Volatile Environments (Hassan & Clack, in Proceedings GECCO 2008, 1507-1514, ACM Press, 2008).

The following invited paper in the journal special issue on “Bio-Inspired Learning and Intelligent Systems” presents a Genetic Programming technique for portfolio optimisation; the first such technique to propose preserving diversity of dynamic phenotypic behaviour, and optimisation of investment returns rather than price-prediction. It is robust to market volatility, reducing down-time while gathering data for retraining, and demonstrates a 3-times performance improvement. The technique is being used by SIAM Capital, is included in Hitoshi Iba's recent book on Evolutionary Computation in Financial Engineering, and has been shown to have superior performance to Support Vector Machines (see above).

  • Evolving robust GP solutions for hedge fund stock selection in emerging markets (Yan & Clack, Soft Computing 15(1):37-50, Springer 2011)