Wavelet-based relevance vector regression model coupled with phase space reconstruction for exchange rate forecasting

Shian Chang Huang, Chia Hsun Hsieh

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Due to the high risk associated with international transactions, exchange rate forecasting is a challenging and important field in modern time series analysis. The difficulty in forecasting arises from the nonlinearity and non-stationarity inherent in exchange rate dynamics. To address these problems, this study proposes a hybrid model that couples two effective feature extraction techniques, phase space reconstruction, and wavelet analysis, with a Relevance Vector Regression (RVR) model to forecast chaotic exchange rates. The time series inputs are first mapped into high-dimension phase space, and then the phase space signal is decomposed on a wavelet basis to analyze its dynamics under various frequencies. Finally, each wavelet component is fed into a local RVR to perform non-parametric regression and forecasting. Compared with other forecasting models, such as support vector machines (SVR), RVR, GJR-GARCH or pure wavelet-base models, the proposed model performs best and statistically improves forecasting performance under root mean square error (RMSE), mean absolute error (MAE) and directional symmetry (DS).

Original languageEnglish
Pages (from-to)1917-1930
Number of pages14
JournalInternational Journal of Innovative Computing, Information and Control
Volume8
Issue number3 A
Publication statusPublished - 2012 Mar 1

Fingerprint

Phase Space Reconstruction
Exchange rate
Forecasting
Regression Model
Wavelets
Wavelet Bases
Phase Space
Regression
Time series analysis
Nonstationarity
Generalized Autoregressive Conditional Heteroscedasticity
Wavelet analysis
Wavelet Analysis
Time Series Analysis
Hybrid Model
Nonparametric Regression
Mean square error
Feature Extraction
Higher Dimensions
Transactions

All Science Journal Classification (ASJC) codes

  • Software
  • Theoretical Computer Science
  • Information Systems
  • Computational Theory and Mathematics

Cite this

@article{32fee81c6c4d4b37a78d7927f4d2c388,
title = "Wavelet-based relevance vector regression model coupled with phase space reconstruction for exchange rate forecasting",
abstract = "Due to the high risk associated with international transactions, exchange rate forecasting is a challenging and important field in modern time series analysis. The difficulty in forecasting arises from the nonlinearity and non-stationarity inherent in exchange rate dynamics. To address these problems, this study proposes a hybrid model that couples two effective feature extraction techniques, phase space reconstruction, and wavelet analysis, with a Relevance Vector Regression (RVR) model to forecast chaotic exchange rates. The time series inputs are first mapped into high-dimension phase space, and then the phase space signal is decomposed on a wavelet basis to analyze its dynamics under various frequencies. Finally, each wavelet component is fed into a local RVR to perform non-parametric regression and forecasting. Compared with other forecasting models, such as support vector machines (SVR), RVR, GJR-GARCH or pure wavelet-base models, the proposed model performs best and statistically improves forecasting performance under root mean square error (RMSE), mean absolute error (MAE) and directional symmetry (DS).",
author = "Huang, {Shian Chang} and Hsieh, {Chia Hsun}",
year = "2012",
month = "3",
day = "1",
language = "English",
volume = "8",
pages = "1917--1930",
journal = "International Journal of Innovative Computing, Information and Control",
issn = "1349-4198",
publisher = "IJICIC Editorial Office",
number = "3 A",

}

TY - JOUR

T1 - Wavelet-based relevance vector regression model coupled with phase space reconstruction for exchange rate forecasting

AU - Huang, Shian Chang

AU - Hsieh, Chia Hsun

PY - 2012/3/1

Y1 - 2012/3/1

N2 - Due to the high risk associated with international transactions, exchange rate forecasting is a challenging and important field in modern time series analysis. The difficulty in forecasting arises from the nonlinearity and non-stationarity inherent in exchange rate dynamics. To address these problems, this study proposes a hybrid model that couples two effective feature extraction techniques, phase space reconstruction, and wavelet analysis, with a Relevance Vector Regression (RVR) model to forecast chaotic exchange rates. The time series inputs are first mapped into high-dimension phase space, and then the phase space signal is decomposed on a wavelet basis to analyze its dynamics under various frequencies. Finally, each wavelet component is fed into a local RVR to perform non-parametric regression and forecasting. Compared with other forecasting models, such as support vector machines (SVR), RVR, GJR-GARCH or pure wavelet-base models, the proposed model performs best and statistically improves forecasting performance under root mean square error (RMSE), mean absolute error (MAE) and directional symmetry (DS).

AB - Due to the high risk associated with international transactions, exchange rate forecasting is a challenging and important field in modern time series analysis. The difficulty in forecasting arises from the nonlinearity and non-stationarity inherent in exchange rate dynamics. To address these problems, this study proposes a hybrid model that couples two effective feature extraction techniques, phase space reconstruction, and wavelet analysis, with a Relevance Vector Regression (RVR) model to forecast chaotic exchange rates. The time series inputs are first mapped into high-dimension phase space, and then the phase space signal is decomposed on a wavelet basis to analyze its dynamics under various frequencies. Finally, each wavelet component is fed into a local RVR to perform non-parametric regression and forecasting. Compared with other forecasting models, such as support vector machines (SVR), RVR, GJR-GARCH or pure wavelet-base models, the proposed model performs best and statistically improves forecasting performance under root mean square error (RMSE), mean absolute error (MAE) and directional symmetry (DS).

UR - http://www.scopus.com/inward/record.url?scp=84863257783&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84863257783&partnerID=8YFLogxK

M3 - Article

AN - SCOPUS:84863257783

VL - 8

SP - 1917

EP - 1930

JO - International Journal of Innovative Computing, Information and Control

JF - International Journal of Innovative Computing, Information and Control

SN - 1349-4198

IS - 3 A

ER -