Genetic programs, data-snooping, and technical analysis, страница 11

While the results of this study do not preclude the existence of profitable technical trading strategies, these results do undermine the previous implications that technical analysis is clearly useful and/or profitable in commodity trading. Instead, these results suggest that rules that are chosen to reduce or eliminate possible sources of data-snooping will likely be less profitable than those chosen otherwise. Further, this study highlights the difficulty in defending ex post the choice of a given technical trading rule. Any technical rule chosen must be ex ante optimal by some objective criteria, or an explanation given as to what constitutes the optimally suboptimal trading rule.As such, for future research into the profitability of technical analysis in commodity markets, the results in TableIVmustrepresenta“lowerbound”fortheprofitsofoptimalin-sample rules in the universe of technical analysis.

Two obvious avenues for the extension of this work exist. The explicit incorporation of various trading rule families (such as momentum, RSI, or others listed in Table II) into the set of function nodes would allow a more direct examination of the trading rules currently in practice. Even though many of these rules are nested within the current framework, including them specifically would allow more direct comparison with technical trading practices as well as making it easier to identify and test optimal trading rules within particular subdomains of technical analysis, allowing for example, the identification and testing of the optimal moving average or filter rule for comparison with the results of Brock et al. (1992). The second direction would be the use of the data and optimality criteria of previous research, such as or Lukac et al. (1988a) to estimate the optimal in- and out-of-sample rules using GP, and comparing the performance of these rules to those analyzed in the original studies.

BIBLIOGRAPHY

Alexander, S. S. (1964). Price movements in speculative markets: Trends or random walks, no. 2. In P. Cootner (Ed.), The random character of stock market changes (pp. 338–372). Cambridge, MA: MIT Press.

Allen, F., & Karjalainen, R. (1999). Using genetic algorithms to find technical trading rules. Journal of Financial Economics, 51, 245–271.

Brock, W., Lakonishok, J., & LeBaron, B. (1992). Simple technical trading rules and the stochastic properties of stock returns. Journal of Finance, 47, 1731–1764.

Brorsen, B. W., & Irwin, S. H. (1987). Futures funds and price volatility. Review of Futures Markets, 7, 118–135.

Campbell, J. Y., Lo, A. W., & MacKinlay, A. C. (1997). The econometrics of financial markets. Princeton, NJ: Princeton University Press.

Cooper, M., & Gulen, H. (2001, November). Is time-series based predictability evident in real-time? Working paper, Purdue University, Krannert School of Management.

Fama, E., & Blume, M. (1966). Filter rules and stock market trading. Journal of Business, 3, 226–241.

Koza, J. R. (1992). Genetic programming: On the programming of computers by means of natural selection. Cambridge, MA: MIT Press.

Lui, Y. H., & Mole, D. (1998). The use of fundamental and technical analyses by foreign exchange dealers: Hong Kong evidence. Journal of International Money and Finance, 17, 535–545.

Lukac, L. P., & Brorsen, B. W. (1990). A comprehensive test of futures market disequilibrium. The Financial Review, 25, 593–622.