Presentation 2008-01-31
Nonlinear Prediction for Top and Bottom Values of Time Series
Masaki OTA, Tomoya SUZUKI,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) To predict stock price and foreign exchange rates, it is efficient to follow only these top and bottom values more than every tick time series from the viewpoint of the number of tradings, especially, for day traders. Top and bottom values of time series have two important information: the timing and the degree of these values. We need to predict the timings and the degrees to make gain. However, because these top and bottom values do not happen frequently, we have to perform long-term prediction. In the case of chaotic time series, it is well known that the prediction errors grow exponentially by long-term prediction. In the present study, we propose a new prediction method for top and bottom values of time series. We decide top and bottom values numerically, and use only these values for every prediction. By this method, we can consider predictions of top and bottom values as short-term predictions. Moreover, we performed numerical simulations to show the validity of our proposed method with chaotic time series and real time series of stock prices and foreign exchange rates.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Economic tick data / Prediction of top and bottom values / Nonlinear prediction / Local linear approximation method
Paper # NLP2007-131
Date of Issue

Conference Information
Committee NLP
Conference Date 2008/1/24(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Nonlinear Problems (NLP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Nonlinear Prediction for Top and Bottom Values of Time Series
Sub Title (in English)
Keyword(1) Economic tick data
Keyword(2) Prediction of top and bottom values
Keyword(3) Nonlinear prediction
Keyword(4) Local linear approximation method
1st Author's Name Masaki OTA
1st Author's Affiliation Doshisha University, Department of Information System Design, Faculty of Engineering()
2nd Author's Name Tomoya SUZUKI
2nd Author's Affiliation Doshisha University, Department of Information System Design, Faculty of Engineering
Date 2008-01-31
Paper # NLP2007-131
Volume (vol) vol.107
Number (no) 477
Page pp.pp.-
#Pages 6
Date of Issue