Developing time-based clustering neural networks to use change-point detection

Application to financial time series

Kyong Joo Oh, Tab Hyup Roh, Myung Sang Moon

Research output: Contribution to journalArticle

7 Citations (Scopus)

Abstract

This study suggests time-based clustering models integrating change-point detection and neural networks, and applies them to financial time series forecasting. The basic concept of the proposed models is to obtain intervals divided by change points, to identify them as change-point groups, and to involve them in the forecasting model. The proposed models consist of two stages. The first stage, the clustering neural network modeling stage, is to detect successive change points in the dataset, and to forecast change-point groups with backpropagation neural networks (BPNs). In this stage, three change-point detection methods are applied and compared. They are: (1) the parametric approach, (2) the nonparametric approach, and (3) the model-based approach. The next stage is to forecast the final output with BPNs. Through the application to financial time series forecasting, we compare the proposed models with a neural network model alone and, in addition, determine which of three change-point detection methods performs better. Furthermore, we evaluate whether the proposed models play a role in clustering to reflect the time. Finally, this study examines the predictability of the integrated neural network models based on change-point detection.

Original languageEnglish
Pages (from-to)51-70
Number of pages20
JournalAsia-Pacific Journal of Operational Research
Volume22
Issue number1
DOIs
Publication statusPublished - 2005 Mar 1

Fingerprint

Change point
Financial time series
Clustering
Neural networks
Time series forecasting
Back-propagation neural network
Network model
Modeling
Time study
Integrated
Predictability

All Science Journal Classification (ASJC) codes

  • Management Science and Operations Research

Cite this

@article{061663c80e4b42f48129e0a0f25353af,
title = "Developing time-based clustering neural networks to use change-point detection: Application to financial time series",
abstract = "This study suggests time-based clustering models integrating change-point detection and neural networks, and applies them to financial time series forecasting. The basic concept of the proposed models is to obtain intervals divided by change points, to identify them as change-point groups, and to involve them in the forecasting model. The proposed models consist of two stages. The first stage, the clustering neural network modeling stage, is to detect successive change points in the dataset, and to forecast change-point groups with backpropagation neural networks (BPNs). In this stage, three change-point detection methods are applied and compared. They are: (1) the parametric approach, (2) the nonparametric approach, and (3) the model-based approach. The next stage is to forecast the final output with BPNs. Through the application to financial time series forecasting, we compare the proposed models with a neural network model alone and, in addition, determine which of three change-point detection methods performs better. Furthermore, we evaluate whether the proposed models play a role in clustering to reflect the time. Finally, this study examines the predictability of the integrated neural network models based on change-point detection.",
author = "Oh, {Kyong Joo} and Roh, {Tab Hyup} and Moon, {Myung Sang}",
year = "2005",
month = "3",
day = "1",
doi = "10.1142/S0217595905000431",
language = "English",
volume = "22",
pages = "51--70",
journal = "Asia-Pacific Journal of Operational Research",
issn = "0217-5959",
publisher = "World Scientific Publishing Co. Pte Ltd",
number = "1",

}

Developing time-based clustering neural networks to use change-point detection : Application to financial time series. / Oh, Kyong Joo; Roh, Tab Hyup; Moon, Myung Sang.

In: Asia-Pacific Journal of Operational Research, Vol. 22, No. 1, 01.03.2005, p. 51-70.

Research output: Contribution to journalArticle

TY - JOUR

T1 - Developing time-based clustering neural networks to use change-point detection

T2 - Application to financial time series

AU - Oh, Kyong Joo

AU - Roh, Tab Hyup

AU - Moon, Myung Sang

PY - 2005/3/1

Y1 - 2005/3/1

N2 - This study suggests time-based clustering models integrating change-point detection and neural networks, and applies them to financial time series forecasting. The basic concept of the proposed models is to obtain intervals divided by change points, to identify them as change-point groups, and to involve them in the forecasting model. The proposed models consist of two stages. The first stage, the clustering neural network modeling stage, is to detect successive change points in the dataset, and to forecast change-point groups with backpropagation neural networks (BPNs). In this stage, three change-point detection methods are applied and compared. They are: (1) the parametric approach, (2) the nonparametric approach, and (3) the model-based approach. The next stage is to forecast the final output with BPNs. Through the application to financial time series forecasting, we compare the proposed models with a neural network model alone and, in addition, determine which of three change-point detection methods performs better. Furthermore, we evaluate whether the proposed models play a role in clustering to reflect the time. Finally, this study examines the predictability of the integrated neural network models based on change-point detection.

AB - This study suggests time-based clustering models integrating change-point detection and neural networks, and applies them to financial time series forecasting. The basic concept of the proposed models is to obtain intervals divided by change points, to identify them as change-point groups, and to involve them in the forecasting model. The proposed models consist of two stages. The first stage, the clustering neural network modeling stage, is to detect successive change points in the dataset, and to forecast change-point groups with backpropagation neural networks (BPNs). In this stage, three change-point detection methods are applied and compared. They are: (1) the parametric approach, (2) the nonparametric approach, and (3) the model-based approach. The next stage is to forecast the final output with BPNs. Through the application to financial time series forecasting, we compare the proposed models with a neural network model alone and, in addition, determine which of three change-point detection methods performs better. Furthermore, we evaluate whether the proposed models play a role in clustering to reflect the time. Finally, this study examines the predictability of the integrated neural network models based on change-point detection.

UR - http://www.scopus.com/inward/record.url?scp=16644401397&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=16644401397&partnerID=8YFLogxK

U2 - 10.1142/S0217595905000431

DO - 10.1142/S0217595905000431

M3 - Article

VL - 22

SP - 51

EP - 70

JO - Asia-Pacific Journal of Operational Research

JF - Asia-Pacific Journal of Operational Research

SN - 0217-5959

IS - 1

ER -