Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping

Thunchira Thongmee, Hiroto Suzuki, Takahiro Ohno, Udom Silparcha

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

    1 Citation (Scopus)

    Abstract

    This paper proposes the Blockwise Strong Relationship (BSR) method that calculates the degree of relationship between any pair of stocks based on only their prices. Our method deploys the data transformation adapted from the symbolic aggregation approximation (SAX) and the distance measure using dynamic time warping (DTW). We propose that the time series data should be processed in blocks of some appropriate size rather than the whole series at once. The experiment was done using IMI Energy indices. The result shows that our method can accurately draw the strongest related pair of stocks out of those that all look related on the surface.

    Original languageEnglish
    Title of host publicationINISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings
    PublisherIEEE Computer Society
    Pages104-109
    Number of pages6
    ISBN (Print)9781479930197
    DOIs
    Publication statusPublished - 2014
    Event2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2014 - Alberobello
    Duration: 2014 Jun 232014 Jun 25

    Other

    Other2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2014
    CityAlberobello
    Period14/6/2314/6/25

    Fingerprint

    Time series
    Agglomeration
    Experiments

    Keywords

    • Dynamic time warping
    • Symbolic representation
    • Time series

    ASJC Scopus subject areas

    • Artificial Intelligence
    • Computer Science Applications

    Cite this

    Thongmee, T., Suzuki, H., Ohno, T., & Silparcha, U. (2014). Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping. In INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings (pp. 104-109). [6873604] IEEE Computer Society. https://doi.org/10.1109/INISTA.2014.6873604

    Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping. / Thongmee, Thunchira; Suzuki, Hiroto; Ohno, Takahiro; Silparcha, Udom.

    INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings. IEEE Computer Society, 2014. p. 104-109 6873604.

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

    Thongmee, T, Suzuki, H, Ohno, T & Silparcha, U 2014, Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping. in INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings., 6873604, IEEE Computer Society, pp. 104-109, 2014 IEEE International Symposium on Innovations in Intelligent Systems and Applications, INISTA 2014, Alberobello, 14/6/23. https://doi.org/10.1109/INISTA.2014.6873604
    Thongmee T, Suzuki H, Ohno T, Silparcha U. Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping. In INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings. IEEE Computer Society. 2014. p. 104-109. 6873604 https://doi.org/10.1109/INISTA.2014.6873604
    Thongmee, Thunchira ; Suzuki, Hiroto ; Ohno, Takahiro ; Silparcha, Udom. / Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping. INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings. IEEE Computer Society, 2014. pp. 104-109
    @inproceedings{39c4a329a0ce49d3bd413fdbe28da3a2,
    title = "Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping",
    abstract = "This paper proposes the Blockwise Strong Relationship (BSR) method that calculates the degree of relationship between any pair of stocks based on only their prices. Our method deploys the data transformation adapted from the symbolic aggregation approximation (SAX) and the distance measure using dynamic time warping (DTW). We propose that the time series data should be processed in blocks of some appropriate size rather than the whole series at once. The experiment was done using IMI Energy indices. The result shows that our method can accurately draw the strongest related pair of stocks out of those that all look related on the surface.",
    keywords = "Dynamic time warping, Symbolic representation, Time series",
    author = "Thunchira Thongmee and Hiroto Suzuki and Takahiro Ohno and Udom Silparcha",
    year = "2014",
    doi = "10.1109/INISTA.2014.6873604",
    language = "English",
    isbn = "9781479930197",
    pages = "104--109",
    booktitle = "INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings",
    publisher = "IEEE Computer Society",

    }

    TY - GEN

    T1 - Finding Strong Relationships of stock prices using blockwise symbolic representation with dynamic time warping

    AU - Thongmee, Thunchira

    AU - Suzuki, Hiroto

    AU - Ohno, Takahiro

    AU - Silparcha, Udom

    PY - 2014

    Y1 - 2014

    N2 - This paper proposes the Blockwise Strong Relationship (BSR) method that calculates the degree of relationship between any pair of stocks based on only their prices. Our method deploys the data transformation adapted from the symbolic aggregation approximation (SAX) and the distance measure using dynamic time warping (DTW). We propose that the time series data should be processed in blocks of some appropriate size rather than the whole series at once. The experiment was done using IMI Energy indices. The result shows that our method can accurately draw the strongest related pair of stocks out of those that all look related on the surface.

    AB - This paper proposes the Blockwise Strong Relationship (BSR) method that calculates the degree of relationship between any pair of stocks based on only their prices. Our method deploys the data transformation adapted from the symbolic aggregation approximation (SAX) and the distance measure using dynamic time warping (DTW). We propose that the time series data should be processed in blocks of some appropriate size rather than the whole series at once. The experiment was done using IMI Energy indices. The result shows that our method can accurately draw the strongest related pair of stocks out of those that all look related on the surface.

    KW - Dynamic time warping

    KW - Symbolic representation

    KW - Time series

    UR - http://www.scopus.com/inward/record.url?scp=84906657338&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=84906657338&partnerID=8YFLogxK

    U2 - 10.1109/INISTA.2014.6873604

    DO - 10.1109/INISTA.2014.6873604

    M3 - Conference contribution

    AN - SCOPUS:84906657338

    SN - 9781479930197

    SP - 104

    EP - 109

    BT - INISTA 2014 - IEEE International Symposium on Innovations in Intelligent Systems and Applications, Proceedings

    PB - IEEE Computer Society

    ER -