Tick-wise motions of foreign exchange rates are shown to be partly predictable, while daily price changes can be approximated by the random walk. We summarize several known facts on the tick-wise price changes in regard to the probability distribution functions, memory-depth of the time series, and so on, and discuss the possibility of predicting the direction of the next tick by using the evolutional computation that leams local values of conditional probabilities from non-stationary environment of the real tick data. © Springer-Verlag 2004.
CITATION STYLE
Tanaka-Yamawaki, M. (2004). Tick-Wise Predictions of Foreign Exchange Rates. Lecture Notes in Computer Science (Including Subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics), 3213, 449–454. https://doi.org/10.1007/978-3-540-30132-5_64
Mendeley helps you to discover research relevant for your work.