Abstract
This paper presents a simple string representation for hourly foreign exchange data and evaluates the performance of a trading strategy derived from it. We make use of a natural discretisation of the time-series based on arbitrary partitioning of the real valued hourly returns to create an alphabet and combine these individual characters to construct a string. The trading decision for each string is learnt in an incremental manner and is thus subject to temporal fluctuations. This naive representation and strategy is compared to the support vector machine, a popular machine learning algorithm for financial time series prediction, that is able to make use of the continuous form of past prices and complex kernel representations. Our extensive experiments show that the simple string representation is capable of outperforming these more exotic approaches, whilst supporting the idea that when it comes to working in high noise environments often the simplest approach is the most effective.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Notes
- 1.
An extension of the original SVM that penalises the slack variables according to their squared value.
- 2.
Technical indicators are best described as rule-based evaluations of the underlying time series where their mathematical formulae is not based on statistical theory but on an expert’s domain knowledge instead.
References
Tay, F.E.H., Cao, L.: Application of support vector machines in financial time series forecasting. Omega 29, 309–317 (2001)
Kim, K.J.: Financial time series forecasting using support vector machines. Neurocomputing 55, 307–319 (2003)
Van Gestel, T., Suykens, J.A.K., Baestaens, D.E., Lambrechts, A., Lanckriet, G., Vandaele, B., De Moor, B., Vandewalle, J.: Financial time series prediction using least squares support vector machines within the evidence framework. Neural Netw. 12, 809–821 (2001)
Cao, L.J., Tay, F.E.H.: Support vector machine with adaptive parameters in financial time series forecasting. Neural Netw. 14, 1506–1518 (2003)
Perez-Cruz, F., Afonso-Rodriguez, J.A., Giner, J.: Estimating GARCH models using support vector machines. Quant. Finan. 3, 163–172 (2003)
Hossain, A., Nasser, M.: Reccurent support and relevance vector machines based model with applications to forecasting volatility of financial returns. J. Intell. Learn. Syst. Appl. 3, 230–241 (2011)
Ou, P., Wang, H.: Financial volatility forecasting by least square support vector machine based on GARCH, EGARCH and GJR models: evidence from ASEAN stock markets. Int. J. Econ. Finan. 2, 51–64 (2010)
Khan, A.I.: Financial volatility forecasting by nonlinear support vector machine heterogeneous autoregressive model: evidence from NIKKEI225 stock index. Int. J. Econ. Finan. 3, 138–150 (2011)
Armstrong, J.S., Adya, M., Collopy, F.: Rule-Based Forecasting: Using Judgment in Time-Series Extrapolation. Kluwer Academic Publishers, Norwell (2001)
Shawe-Taylor, J., Cristianini, N.: Kernel Methods for Pattern Analysis Infrastructure. Cambridge University Press, New York (2004)
Lodhi, H., Saunders, C., Shawe-Taylor, J., Cristianini, N., Watkins, C.: Text classification using string kernels. J. Mach. Learn. Res. 2, 419–444 (2002)
Liu, B., Wang, X., Lin, L., Dong, Q., Wang, X.: A discriminative method for protein remote homology detection and fold recognition combining top-n-grams and latent semantic analysis. BMC Bioinf. 9, 510 (2008)
Author information
Authors and Affiliations
Corresponding authors
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2016 Springer International Publishing Switzerland
About this paper
Cite this paper
Cousins, S., Žličar, B. (2016). Evaluating a Simple String Representation for Intra-day Foreign Exchange Prediction. In: Ceci, M., Loglisci, C., Manco, G., Masciari, E., Ras, Z. (eds) New Frontiers in Mining Complex Patterns. NFMCP 2015. Lecture Notes in Computer Science(), vol 9607. Springer, Cham. https://doi.org/10.1007/978-3-319-39315-5_15
Download citation
DOI: https://doi.org/10.1007/978-3-319-39315-5_15
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-39314-8
Online ISBN: 978-3-319-39315-5
eBook Packages: Computer ScienceComputer Science (R0)