Detecting precursory events in time series data by an extension of singular spectrum transformation

Terumasa Tokunaga, Daisuke Ikeda, Kazuyuki Nakamura, Tomoyuki Higuchi, Akimasa Yoshikawa, Teiji Uozumi, Akiko Fujimoto, Akira Morioka, Kiyofumi Yumoto

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Abstract

To predict an occurrence of extraordinary phenomena, such as earthquakes, failures of engineering system and financial market crushes, it is important to identify precursory events in time series. However, existing methods are limited in their applicability for real world precursor detections. Recently, Ide and Inoue [1] have developed an SSA-based change-point detection method, called singular spectrum transformation (SST). In this paper, we extend the SST so that it is applicable for real world precursor detections, focusing on the wide applicability of the conventional SST. Although the SST is suitable for detecting various types of change-points, detecting precursors can be far more difficult than expected because, in general, real world time series contains measurement noise and non-stationary trends. Furthermore, precursory events are usually observed as minute and less-visible fluctuations preceding an onset of massive fluctuations of extraordinary phenomena and therefore they are easily over-looked. To overcome this, we extend the conventional SST to the multivariable SST, focusing on the synchronism detection of precursory events in multiple sequences of univariate time series. First, we would like to define the problem setting of real world precursory detections and consider its difficulties. Second, the multivariable SST is introduced. Third, we apply SST to geomagnetic time series data and show the multivariable SST is more suitable than the conventional SST for real world precursor detections. Finally, we show further experimental results using artificial data to evaluate the reliability of SST-based precursor detections.

Original languageEnglish
Title of host publicationSelected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10
Pages366-374
Number of pages9
Publication statusPublished - Dec 1 2010
Event10th WSEAS International Conference on Applied Computer Science, ACS'10 - Iwate, Japan
Duration: Oct 4 2010Oct 6 2010

Publication series

NameInternational Conference on Applied Computer Science - Proceedings
ISSN (Print)1792-4863

Other

Other10th WSEAS International Conference on Applied Computer Science, ACS'10
CountryJapan
CityIwate
Period10/4/1010/6/10

Fingerprint

Time series
Systems engineering
Earthquakes
Synchronization

All Science Journal Classification (ASJC) codes

  • Computer Science (miscellaneous)

Cite this

Tokunaga, T., Ikeda, D., Nakamura, K., Higuchi, T., Yoshikawa, A., Uozumi, T., ... Yumoto, K. (2010). Detecting precursory events in time series data by an extension of singular spectrum transformation. In Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10 (pp. 366-374). (International Conference on Applied Computer Science - Proceedings).

Detecting precursory events in time series data by an extension of singular spectrum transformation. / Tokunaga, Terumasa; Ikeda, Daisuke; Nakamura, Kazuyuki; Higuchi, Tomoyuki; Yoshikawa, Akimasa; Uozumi, Teiji; Fujimoto, Akiko; Morioka, Akira; Yumoto, Kiyofumi.

Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10. 2010. p. 366-374 (International Conference on Applied Computer Science - Proceedings).

Research output: Chapter in Book/Report/Conference proceedingConference contribution

Tokunaga, T, Ikeda, D, Nakamura, K, Higuchi, T, Yoshikawa, A, Uozumi, T, Fujimoto, A, Morioka, A & Yumoto, K 2010, Detecting precursory events in time series data by an extension of singular spectrum transformation. in Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10. International Conference on Applied Computer Science - Proceedings, pp. 366-374, 10th WSEAS International Conference on Applied Computer Science, ACS'10, Iwate, Japan, 10/4/10.
Tokunaga T, Ikeda D, Nakamura K, Higuchi T, Yoshikawa A, Uozumi T et al. Detecting precursory events in time series data by an extension of singular spectrum transformation. In Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10. 2010. p. 366-374. (International Conference on Applied Computer Science - Proceedings).
Tokunaga, Terumasa ; Ikeda, Daisuke ; Nakamura, Kazuyuki ; Higuchi, Tomoyuki ; Yoshikawa, Akimasa ; Uozumi, Teiji ; Fujimoto, Akiko ; Morioka, Akira ; Yumoto, Kiyofumi. / Detecting precursory events in time series data by an extension of singular spectrum transformation. Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10. 2010. pp. 366-374 (International Conference on Applied Computer Science - Proceedings).
@inproceedings{b4c1f344320745f58a24973279de5047,
title = "Detecting precursory events in time series data by an extension of singular spectrum transformation",
abstract = "To predict an occurrence of extraordinary phenomena, such as earthquakes, failures of engineering system and financial market crushes, it is important to identify precursory events in time series. However, existing methods are limited in their applicability for real world precursor detections. Recently, Ide and Inoue [1] have developed an SSA-based change-point detection method, called singular spectrum transformation (SST). In this paper, we extend the SST so that it is applicable for real world precursor detections, focusing on the wide applicability of the conventional SST. Although the SST is suitable for detecting various types of change-points, detecting precursors can be far more difficult than expected because, in general, real world time series contains measurement noise and non-stationary trends. Furthermore, precursory events are usually observed as minute and less-visible fluctuations preceding an onset of massive fluctuations of extraordinary phenomena and therefore they are easily over-looked. To overcome this, we extend the conventional SST to the multivariable SST, focusing on the synchronism detection of precursory events in multiple sequences of univariate time series. First, we would like to define the problem setting of real world precursory detections and consider its difficulties. Second, the multivariable SST is introduced. Third, we apply SST to geomagnetic time series data and show the multivariable SST is more suitable than the conventional SST for real world precursor detections. Finally, we show further experimental results using artificial data to evaluate the reliability of SST-based precursor detections.",
author = "Terumasa Tokunaga and Daisuke Ikeda and Kazuyuki Nakamura and Tomoyuki Higuchi and Akimasa Yoshikawa and Teiji Uozumi and Akiko Fujimoto and Akira Morioka and Kiyofumi Yumoto",
year = "2010",
month = "12",
day = "1",
language = "English",
isbn = "9789604742318",
series = "International Conference on Applied Computer Science - Proceedings",
pages = "366--374",
booktitle = "Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10",

}

TY - GEN

T1 - Detecting precursory events in time series data by an extension of singular spectrum transformation

AU - Tokunaga, Terumasa

AU - Ikeda, Daisuke

AU - Nakamura, Kazuyuki

AU - Higuchi, Tomoyuki

AU - Yoshikawa, Akimasa

AU - Uozumi, Teiji

AU - Fujimoto, Akiko

AU - Morioka, Akira

AU - Yumoto, Kiyofumi

PY - 2010/12/1

Y1 - 2010/12/1

N2 - To predict an occurrence of extraordinary phenomena, such as earthquakes, failures of engineering system and financial market crushes, it is important to identify precursory events in time series. However, existing methods are limited in their applicability for real world precursor detections. Recently, Ide and Inoue [1] have developed an SSA-based change-point detection method, called singular spectrum transformation (SST). In this paper, we extend the SST so that it is applicable for real world precursor detections, focusing on the wide applicability of the conventional SST. Although the SST is suitable for detecting various types of change-points, detecting precursors can be far more difficult than expected because, in general, real world time series contains measurement noise and non-stationary trends. Furthermore, precursory events are usually observed as minute and less-visible fluctuations preceding an onset of massive fluctuations of extraordinary phenomena and therefore they are easily over-looked. To overcome this, we extend the conventional SST to the multivariable SST, focusing on the synchronism detection of precursory events in multiple sequences of univariate time series. First, we would like to define the problem setting of real world precursory detections and consider its difficulties. Second, the multivariable SST is introduced. Third, we apply SST to geomagnetic time series data and show the multivariable SST is more suitable than the conventional SST for real world precursor detections. Finally, we show further experimental results using artificial data to evaluate the reliability of SST-based precursor detections.

AB - To predict an occurrence of extraordinary phenomena, such as earthquakes, failures of engineering system and financial market crushes, it is important to identify precursory events in time series. However, existing methods are limited in their applicability for real world precursor detections. Recently, Ide and Inoue [1] have developed an SSA-based change-point detection method, called singular spectrum transformation (SST). In this paper, we extend the SST so that it is applicable for real world precursor detections, focusing on the wide applicability of the conventional SST. Although the SST is suitable for detecting various types of change-points, detecting precursors can be far more difficult than expected because, in general, real world time series contains measurement noise and non-stationary trends. Furthermore, precursory events are usually observed as minute and less-visible fluctuations preceding an onset of massive fluctuations of extraordinary phenomena and therefore they are easily over-looked. To overcome this, we extend the conventional SST to the multivariable SST, focusing on the synchronism detection of precursory events in multiple sequences of univariate time series. First, we would like to define the problem setting of real world precursory detections and consider its difficulties. Second, the multivariable SST is introduced. Third, we apply SST to geomagnetic time series data and show the multivariable SST is more suitable than the conventional SST for real world precursor detections. Finally, we show further experimental results using artificial data to evaluate the reliability of SST-based precursor detections.

UR - http://www.scopus.com/inward/record.url?scp=79958702189&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=79958702189&partnerID=8YFLogxK

M3 - Conference contribution

SN - 9789604742318

T3 - International Conference on Applied Computer Science - Proceedings

SP - 366

EP - 374

BT - Selected Topics in Applied Computer Science - 10th WSEAS International Conference on Applied Computer Science, ACS'10

ER -