Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Traditional forecasting models are not very effective in most financial time series. To address the problem, this study proposes a novel system for financial modeling and forecasting. In the first stage, wavelet analysis transforms the input space of raw data to a time-scale feature space suitable for financial modeling and forecasting. A spectral clustering algorithm is then used to partition the feature space into several disjointed regions according to their time series dynamics. In the second stage, multiple kernel partial least square regressors ideally suited to each partitioned region are constructed for final forecasting. The proposed model outperforms neural networks, SVMs, and traditional GARCH models, significantly reducing root-mean-squared forecasting errors.

Original languageEnglish
Pages (from-to)6755-6764
Number of pages10
JournalApplied Mathematics and Computation
Volume217
Issue number15
DOIs
Publication statusPublished - 2011 Apr 1

Fingerprint

Financial Modeling
Spectral Clustering
Partial Least Squares Regression
Forecasting
Wavelets
kernel
Feature Space
Time series
GARCH Model
Financial Time Series
Scale Space
Wavelet analysis
Wavelet Analysis
Partial Least Squares
Clustering algorithms
Neural Network Model
Clustering Algorithm
Time Scales
Partition
Roots

All Science Journal Classification (ASJC) codes

  • Computational Mathematics
  • Applied Mathematics

Cite this

@article{3b925f546a984e9484f80e445ceabf54,
title = "Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting",
abstract = "Traditional forecasting models are not very effective in most financial time series. To address the problem, this study proposes a novel system for financial modeling and forecasting. In the first stage, wavelet analysis transforms the input space of raw data to a time-scale feature space suitable for financial modeling and forecasting. A spectral clustering algorithm is then used to partition the feature space into several disjointed regions according to their time series dynamics. In the second stage, multiple kernel partial least square regressors ideally suited to each partitioned region are constructed for final forecasting. The proposed model outperforms neural networks, SVMs, and traditional GARCH models, significantly reducing root-mean-squared forecasting errors.",
author = "Huang, {Shian Chang}",
year = "2011",
month = "4",
day = "1",
doi = "10.1016/j.amc.2011.01.096",
language = "English",
volume = "217",
pages = "6755--6764",
journal = "Applied Mathematics and Computation",
issn = "0096-3003",
publisher = "Elsevier Inc.",
number = "15",

}

TY - JOUR

T1 - Integrating spectral clustering with wavelet based kernel partial least square regressions for financial modeling and forecasting

AU - Huang, Shian Chang

PY - 2011/4/1

Y1 - 2011/4/1

N2 - Traditional forecasting models are not very effective in most financial time series. To address the problem, this study proposes a novel system for financial modeling and forecasting. In the first stage, wavelet analysis transforms the input space of raw data to a time-scale feature space suitable for financial modeling and forecasting. A spectral clustering algorithm is then used to partition the feature space into several disjointed regions according to their time series dynamics. In the second stage, multiple kernel partial least square regressors ideally suited to each partitioned region are constructed for final forecasting. The proposed model outperforms neural networks, SVMs, and traditional GARCH models, significantly reducing root-mean-squared forecasting errors.

AB - Traditional forecasting models are not very effective in most financial time series. To address the problem, this study proposes a novel system for financial modeling and forecasting. In the first stage, wavelet analysis transforms the input space of raw data to a time-scale feature space suitable for financial modeling and forecasting. A spectral clustering algorithm is then used to partition the feature space into several disjointed regions according to their time series dynamics. In the second stage, multiple kernel partial least square regressors ideally suited to each partitioned region are constructed for final forecasting. The proposed model outperforms neural networks, SVMs, and traditional GARCH models, significantly reducing root-mean-squared forecasting errors.

UR - http://www.scopus.com/inward/record.url?scp=79952364639&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79952364639&partnerID=8YFLogxK

U2 - 10.1016/j.amc.2011.01.096

DO - 10.1016/j.amc.2011.01.096

M3 - Article

AN - SCOPUS:79952364639

VL - 217

SP - 6755

EP - 6764

JO - Applied Mathematics and Computation

JF - Applied Mathematics and Computation

SN - 0096-3003

IS - 15

ER -