Presentation 1997/1/23
A Prediction Method for Multi-dimensional Fractal Signal by Using Scale Expansion
Kazuyoshi IKEDA, Shozo TOKINAGA, Akio MIYAZAKI, Nobuyuki Shimazu,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) This report deals with the prediction method for multi-dimensional signal bearing fractal geometry. The prediction error and the estimation of parameters are also discussed. At first, multi-dimensional signal is represented by the convolution of the impulse responce expanded by a set of scaling function and the input signal. Then, by using the scale expansion for each axis the future signal are estimated. As a result, it, is shown that 1-step ahead prediction is very well, and n-step ahead prediction where forecast, is used as an observation is limited in a range. The parameters such as the fractal dimension are estimeted by using the wavelet coefficients of the signal. Application for the portfolio management and the segmentation of the texture are shown.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) Fractal / Multi-dimension / Prediction / Fractal dimension / stock price
Paper # DSP96-111,SAT96-106,RCS96-117
Date of Issue

Conference Information
Committee DSP
Conference Date 1997/1/23(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 Digital Signal Processing (DSP)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) A Prediction Method for Multi-dimensional Fractal Signal by Using Scale Expansion
Sub Title (in English)
Keyword(1) Fractal
Keyword(2) Multi-dimension
Keyword(3) Prediction
Keyword(4) Fractal dimension
Keyword(5) stock price
1st Author's Name Kazuyoshi IKEDA
1st Author's Affiliation Faculty of Economics, Kyushu University()
2nd Author's Name Shozo TOKINAGA
2nd Author's Affiliation Faculty of Economics, Kyushu University
3rd Author's Name Akio MIYAZAKI
3rd Author's Affiliation Graduate School of Information and System Sciences, Kyushu University
4th Author's Name Nobuyuki Shimazu
4th Author's Affiliation Graduate School of Information and System Sciences, Kyushu University
Date 1997/1/23
Paper # DSP96-111,SAT96-106,RCS96-117
Volume (vol) vol.96
Number (no) 477
Page pp.pp.-
#Pages 7
Date of Issue