Bayesian Nonparametric Regression Modeling of Panel Data for Sequential Classification

Sihan Xiong, Yiwei Fu, Asok Ray

    Research output: Contribution to journalArticle

    Abstract

    This paper proposes a Bayesian nonparametric regression model of panel data for sequential pattern classification. The proposed method provides a flexible and parsimonious model that allows both time-independent spatial variables and time-dependent exogenous variables to be predictors. Not only this method improves the accuracy of parameter estimation for limited data, but also it facilitates model interpretation by identifying statistically significant predictors with hypothesis testing. Moreover, as the data length approaches infinity, posterior consistency of the model is guaranteed for general data-generating processes under regular conditions. The resulting model of panel data can also be used for sequential classification. The proposed method has been tested by numerical simulation, then validated on an econometric public data set, and subsequently validated for detection of combustion instabilities with experimental data that have been generated in a laboratory environment.

    Original languageEnglish (US)
    Article number8066450
    Pages (from-to)4128-4139
    Number of pages12
    JournalIEEE Transactions on Neural Networks and Learning Systems
    Volume29
    Issue number9
    DOIs
    StatePublished - Sep 1 2018

    Fingerprint

    Parameter estimation
    Pattern recognition
    Computer simulation
    Testing

    All Science Journal Classification (ASJC) codes

    • Software
    • Computer Science Applications
    • Computer Networks and Communications
    • Artificial Intelligence

    Cite this

    @article{2bc9857fb0b640bdb4da4bc239982e22,
    title = "Bayesian Nonparametric Regression Modeling of Panel Data for Sequential Classification",
    abstract = "This paper proposes a Bayesian nonparametric regression model of panel data for sequential pattern classification. The proposed method provides a flexible and parsimonious model that allows both time-independent spatial variables and time-dependent exogenous variables to be predictors. Not only this method improves the accuracy of parameter estimation for limited data, but also it facilitates model interpretation by identifying statistically significant predictors with hypothesis testing. Moreover, as the data length approaches infinity, posterior consistency of the model is guaranteed for general data-generating processes under regular conditions. The resulting model of panel data can also be used for sequential classification. The proposed method has been tested by numerical simulation, then validated on an econometric public data set, and subsequently validated for detection of combustion instabilities with experimental data that have been generated in a laboratory environment.",
    author = "Sihan Xiong and Yiwei Fu and Asok Ray",
    year = "2018",
    month = "9",
    day = "1",
    doi = "10.1109/TNNLS.2017.2752005",
    language = "English (US)",
    volume = "29",
    pages = "4128--4139",
    journal = "IEEE Transactions on Neural Networks and Learning Systems",
    issn = "2162-237X",
    publisher = "IEEE Computational Intelligence Society",
    number = "9",

    }

    Bayesian Nonparametric Regression Modeling of Panel Data for Sequential Classification. / Xiong, Sihan; Fu, Yiwei; Ray, Asok.

    In: IEEE Transactions on Neural Networks and Learning Systems, Vol. 29, No. 9, 8066450, 01.09.2018, p. 4128-4139.

    Research output: Contribution to journalArticle

    TY - JOUR

    T1 - Bayesian Nonparametric Regression Modeling of Panel Data for Sequential Classification

    AU - Xiong, Sihan

    AU - Fu, Yiwei

    AU - Ray, Asok

    PY - 2018/9/1

    Y1 - 2018/9/1

    N2 - This paper proposes a Bayesian nonparametric regression model of panel data for sequential pattern classification. The proposed method provides a flexible and parsimonious model that allows both time-independent spatial variables and time-dependent exogenous variables to be predictors. Not only this method improves the accuracy of parameter estimation for limited data, but also it facilitates model interpretation by identifying statistically significant predictors with hypothesis testing. Moreover, as the data length approaches infinity, posterior consistency of the model is guaranteed for general data-generating processes under regular conditions. The resulting model of panel data can also be used for sequential classification. The proposed method has been tested by numerical simulation, then validated on an econometric public data set, and subsequently validated for detection of combustion instabilities with experimental data that have been generated in a laboratory environment.

    AB - This paper proposes a Bayesian nonparametric regression model of panel data for sequential pattern classification. The proposed method provides a flexible and parsimonious model that allows both time-independent spatial variables and time-dependent exogenous variables to be predictors. Not only this method improves the accuracy of parameter estimation for limited data, but also it facilitates model interpretation by identifying statistically significant predictors with hypothesis testing. Moreover, as the data length approaches infinity, posterior consistency of the model is guaranteed for general data-generating processes under regular conditions. The resulting model of panel data can also be used for sequential classification. The proposed method has been tested by numerical simulation, then validated on an econometric public data set, and subsequently validated for detection of combustion instabilities with experimental data that have been generated in a laboratory environment.

    UR - http://www.scopus.com/inward/record.url?scp=85052685288&partnerID=8YFLogxK

    UR - http://www.scopus.com/inward/citedby.url?scp=85052685288&partnerID=8YFLogxK

    U2 - 10.1109/TNNLS.2017.2752005

    DO - 10.1109/TNNLS.2017.2752005

    M3 - Article

    C2 - 29035227

    AN - SCOPUS:85052685288

    VL - 29

    SP - 4128

    EP - 4139

    JO - IEEE Transactions on Neural Networks and Learning Systems

    JF - IEEE Transactions on Neural Networks and Learning Systems

    SN - 2162-237X

    IS - 9

    M1 - 8066450

    ER -