7 CONCLUSION
In this paper, we reviewed a formulation for the mar-
ket timing problem and introduced three new contri-
butions: (i) using trend representative testing to ex-
pose potential solutions to various market conditions
while training and testing; (ii) designed GA and PSO
algorithms to tackle market timing and ﬁnally (iii)
compared our proposed GA and PSO algorithms us-
ing 30 strands (stocks undergoing a particular trend)
and 63 signal generating components. Our results
showed that the PSO variants are competitive to GA –
which is the most widely used metaheuristic in mar-
ket timing – and ranked better when it came to perfor-
mance, with one variant doing so at a fraction of the
number of iterations used by GA.
We suggest the following avenues of future re-
search. First, use a more sophisticated measure of
ﬁnancial ﬁtness. This would allow us to simulate
hidden costs of trading such as slippage. Second,
approach the problem of market timing as a multi-
objective one by trying to maximize performance
across the three types of trends and against multiple ﬁ-
nancial objectives. Finally, adapt more metaheuristics
to tackle market timing and compare its performance
against the currently proposed ones in signiﬁcantly
larger datasets. We could then use meta-learning to
understand if and when metaheuristics perform sig-
niﬁcantly better than others under particular condi-
tions and use that information to build hybrid ap-
proaches that use more than one metaheuristic to build
strategies for market timing.
REFERENCES
Abdelbar, A. (2008). Stubborn ants. In IEEE Swarm Intel-
ligence Symposium, SIS 2008, pages 1 – 5.
Allen, F. and Karjalainen, R. (1999). Using genetic algo-
rithms to ﬁnd technical trading rules. Journal of Fi-
nancial Economics, 51(2):245–271.
Bera, A., Sychel, D., and Sacharski, B. (2014). Improved
Particle Swarm Optimization method for investment
strategies parameters computing. Journal of Theoret-
ical and Applied Computer Science, 8(4):45–55.
Briza, A. C. and Naval Jr., P. C. (2011). Stock trading sys-
tem based on the multi-objective particle swarm opti-
mization of technical indicators on end-of-day market
data. Applied Soft Computing, 11(1):1191–1201.
Chakravarty, S. and Dash, P. K. (2012). A PSO based in-
tegrated functional link net and interval type-2 fuzzy
logic system for predicting stock market indices. Ap-
plied Soft Computing, 12(2):931–941.
Chen, S.-M. and Kao, P.-Y. (2013). TAIEX forecasting
based on fuzzy time series, particle swarm optimiza-
tion techniques and support vector machines. Infor-
mation Sciences, 247:62–71.
Clerc, M. (1999). The swarm and the queen: towards a de-
terministic and adaptive particle swarm optimization.
In Proceedings of the 1999 Congress on Evolution-
ary Computation-CEC99 (Cat. No. 99TH8406), vol-
ume 3, pages 1951–1957.
Clerc, M. (2002). Think locally act locally-a framework for
adaptive particle swarm optimizers. IEEE Journal of
Evolutionary Computation, 29:1951–1957.
de la Fuente, D., Garrido, A., Laviada, J., and G
´
omez, A.
(2006). Genetic algorithms to optimise the time to
make stock market investment. In Genetic and Evolu-
tionary Computation Conference, pages 1857–1858.
Engelbrecht, A. P. (2005). Fundamentals of Computational
Swarm Intelligence. John Wiley & Sons Ltd.
Garc
´
ıa, S., Fern
´
andez, A., Luengo, J., and Herrera, F.
(2010). Advanced nonparametric tests for multiple
comparisons in the design of experiments in compu-
tational intelligence and data mining: Experimental
analysis of power. Inf. Sci., 180(10):2044–2064.
Hu, Y., Liu, K., Zhang, X., Su, L., Ngai, E. W. T., and Liu,
M. (2015). Application of evolutionary computation
for rule discovery in stock algorithmic trading: A lit-
erature review. Applied Soft Computing, 36:534–551.
Kampouridis, M. and Otero, F. E. (2017). Evolving trading
strategies using directional changes. Expert Systems
with Applications, 73:145–160.
Karathanasopoulos, A., Dunis, C., and Khalil, S. (2016).
Modelling, forecasting and trading with a new sliding
window approach: the crack spread example. Quanti-
tative Finance, 7688(September):1–12.
Kaufman, P. J. (2013). Trading Systems and Methods. John
Wiley & Sons, Inc, 5th edition.
Kim, Y., Ahn, W., Oh, K. J., and Enke, D. (2017). An
intelligent hybrid trading system for discovering trad-
ing rules for the futures market using rough sets and
genetic algorithms. Applied Soft Computing, 55:127–
140.
Ladyzynski, P. and Grzegorzewski, P. (2013). Particle
swarm intelligence tunning of fuzzy geometric proto-
forms for price patterns recognition and stock trading.
Expert Systems with Applications, 40(7):2391–2397.
Liu, C. F., Yeh, C. Y., and Lee, S. J. (2012). Application
of type-2 neuro-fuzzy modeling in stock price predic-
tion. Applied Soft Computing, 12(4):1348–1358.
L
´
opez-Ib
´
a
˜
nez, M., Dubois-Lacoste, J., P
´
erez C
´
aceres, L.,
St
¨
utzle, T., and Birattari, M. (2016). The irace pack-
age: Iterated racing for automatic algorithm conﬁgu-
ration. Operations Research Perspectives, 3:43–58.
Mohamed, I. and Otero, F. E. (2018). Using Particle
Swarms to Build Strategies for Market Timing: A
Comparative Study. In Swarm Intelligence: 11th
International Conference, ANTS 2018, Rome, Italy,
October 29–31, 2018, Proceedings, pages 435–436.
Springer International Publishing.
Patterson, S. (2013). Dark Pools: The Rise of A.I. Trad-
ing Machines and the Looming Threat to Wall Street.
Random House Business Books.
ECTA 2019 - 11th International Conference on Evolutionary Computation Theory and Applications
68