Abstract
This chapter proposes and analyzes a new method for predicting the direction of a timeseries, that is, the relative position of future observations with respect to past coordinates, a problem of obvious interest to financial forecasters. The method involves two steps: an embedding step from real-valued observations to discrete values and a prediction step based on statistical inference. Both of these are explained in detail and rigorously justified. Finally, the method is applied to two illustrative time series: the daily closing prices of the S&P500 market index (for the period 1995–2001) and the quarterly growth rates for the US gross domestic product from 1959 till 2000. The results obtained for these two cases are extremely encouraging.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Similar content being viewed by others
References
H. Akaike: `Information Theory and an Extension of the Maximum Likelihood Principle’. In: 2nd International Symposium on Information Theory, ed. by B.N. Petrov and F. Csaki ( Akademiai Kiado, Budapest, 1973 )
D.R. Cox: Biometrika 62, 441 (1975)
F.X. Diebold, J.H. Lee, G.C. Weinbach, ‘Regime Switching with Time Varying Probabilities’. In: Nonstationary Time Series and Cointegration, ed. by C. Hargreaves ( Oxford University Press, Oxford 1994 )
L. Fahrmeir, H. Kaufmann: J. Time Series Analys. 8, 147 (1987)
K. Fokianos and B. Kedem: J. Mult. Analys. 67, 277 (1998)
W. Greene: Econometric Analysis, 4th ed., ( Prentice Hall, Upper Saddle River 2000 )
J.D. Hamilton: Econometrica 57, 357 (1989)
J.D. Hamilton: J. Econometrics 45, 39 (1990)
J.D. Hamilton: J. Bus. Econ. Statist. 9, 27 (1991)
H. Kaufmann. Ann. Statist. 15, 79 (1987)
P. McCullagh, J.A. Neider: Generalized Linear Models, 2nd. ed. ( Chapman and Hall, London 1989 )
F. Schwarz: Ann. Statist. 6, 461 (1978)
E. Slud: Scand. J. Statist. 19, 97 (1992)
W.H. Wong: Ann. Statist. 14, 88 (1986)
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2004 Springer-Verlag Berlin Heidelberg
About this chapter
Cite this chapter
Thomakos, D.D. (2004). Predicting the Direction of a Time Series. In: Wille, L.T. (eds) New Directions in Statistical Physics. Springer, Berlin, Heidelberg. https://doi.org/10.1007/978-3-662-08968-2_1
Download citation
DOI: https://doi.org/10.1007/978-3-662-08968-2_1
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-642-07739-5
Online ISBN: 978-3-662-08968-2
eBook Packages: Springer Book Archive