Genetic Programming in Statistical Arbitrage
This paper employs genetic programming to discover statistical arbitrage strategies on the banking sector in the Euro Stoxx universe. Binary decision rules are evolved using two different representations. The first is the classical single tree approach, while the second is a dual tree structure where evaluation is contingent on the current market position. Hence, buy and sell rules are co-evolved. Both methods are capable of discovering significant statistical arbitrage strategies.
KeywordsGenetic Programming Sharpe Ratio Statistical Arbitrage Trading Rule Dual Tree
Unable to display preview. Download preview PDF.
- 2.Arthur, B., Holland, J.H., LeBaron, B., Palmer, R., Tayler, P.: Asset pricing under endogenous expectations in an artificial stock market, Technical report, Santa Fe Institute (1996)Google Scholar
- 3.Barberis, N., Thaler, R.: Handbook of the Economics of Finance, pp. 1052–1090. Elsevier Science, Amsterdam (2003)Google Scholar
- 4.Becker, L.A., Seshadri, M.: Cooperative coevolution of technical trading rules, Technical report, Department of Computer Science, Worcester Polytechnic Institute (2003a)Google Scholar
- 5.Becker, L.A., Seshadri, M.: GP–evolved technical trading rules can outperform buy and hold, Technical report, Department of Computer Science, Worcester Polytechnic Institute (2003b)Google Scholar
- 11.Jain, P.C., Joh, G.-H.: The dependence between hourly prices and trading volume. Journal of Financial Economics 23(3), 269–283 (1988)Google Scholar
- 14.Jonsson, H., Madjidi, P., Nordahl, M.G.: Evolution of trading rules for the FX market or how to make money out of GP, Technical report, Institute of Theoretical Physics, Chalmers University of Technology (1997)Google Scholar
- 16.Li, J.: FGP: a genetic programming based tool for financial forecasting, PhD thesis, University of Essex (2001)Google Scholar