The Difference, System and ‘Double-D’ GMM Panel Estimators in the Presence of Structural Breaks

Rosen Azad Chowdhury, Bill Russell

    Research output: Contribution to journalArticle

    1 Citation (Scopus)
    78 Downloads (Pure)

    Abstract

    The effects of structural breaks in dynamic panels are more complicated than in time series models as the bias can be either negative or positive. This paper focuses on the effects of mean shifts in otherwise stationary processes within an instrumental variable panel estimation framework. We show the sources of the bias and a Monte Carlo analysis calibrated on United States bank lending data demonstrates the size of the bias for a range of auto-regressive parameters. We also propose additional moment conditions that can be used to reduce the biases caused by shifts in the mean of the data.
    Original languageEnglish
    JournalScottish Journal of Political Economy
    Early online date10 Oct 2017
    DOIs
    Publication statusE-pub ahead of print - 10 Oct 2017

    Fingerprint

    trend
    lending
    time series
    bank
    Structural breaks
    Estimator
    Moment conditions
    Monte Carlo analysis
    Dynamic panel
    Panel estimation
    Instrumental variables
    Bank lending
    Stationary process
    Mean shift
    Time series models

    Keywords

    • Dynamic panel estimators
    • mean shifts/structural breaks
    • bank lending channel

    Cite this

    @article{e6f91170aa9546a783f45afcdf997acf,
    title = "The Difference, System and ‘Double-D’ GMM Panel Estimators in the Presence of Structural Breaks",
    abstract = "The effects of structural breaks in dynamic panels are more complicated than in time series models as the bias can be either negative or positive. This paper focuses on the effects of mean shifts in otherwise stationary processes within an instrumental variable panel estimation framework. We show the sources of the bias and a Monte Carlo analysis calibrated on United States bank lending data demonstrates the size of the bias for a range of auto-regressive parameters. We also propose additional moment conditions that can be used to reduce the biases caused by shifts in the mean of the data.",
    keywords = "Dynamic panel estimators, mean shifts/structural breaks, bank lending channel",
    author = "Chowdhury, {Rosen Azad} and Bill Russell",
    note = "{\circledC} 2017 Scottish Economic Society",
    year = "2017",
    month = "10",
    day = "10",
    doi = "10.1111/sjpe.12142",
    language = "English",
    journal = "Scottish Journal of Political Economy",
    issn = "0036-9292",
    publisher = "Wiley",

    }

    TY - JOUR

    T1 - The Difference, System and ‘Double-D’ GMM Panel Estimators in the Presence of Structural Breaks

    AU - Chowdhury, Rosen Azad

    AU - Russell, Bill

    N1 - © 2017 Scottish Economic Society

    PY - 2017/10/10

    Y1 - 2017/10/10

    N2 - The effects of structural breaks in dynamic panels are more complicated than in time series models as the bias can be either negative or positive. This paper focuses on the effects of mean shifts in otherwise stationary processes within an instrumental variable panel estimation framework. We show the sources of the bias and a Monte Carlo analysis calibrated on United States bank lending data demonstrates the size of the bias for a range of auto-regressive parameters. We also propose additional moment conditions that can be used to reduce the biases caused by shifts in the mean of the data.

    AB - The effects of structural breaks in dynamic panels are more complicated than in time series models as the bias can be either negative or positive. This paper focuses on the effects of mean shifts in otherwise stationary processes within an instrumental variable panel estimation framework. We show the sources of the bias and a Monte Carlo analysis calibrated on United States bank lending data demonstrates the size of the bias for a range of auto-regressive parameters. We also propose additional moment conditions that can be used to reduce the biases caused by shifts in the mean of the data.

    KW - Dynamic panel estimators

    KW - mean shifts/structural breaks

    KW - bank lending channel

    U2 - 10.1111/sjpe.12142

    DO - 10.1111/sjpe.12142

    M3 - Article

    JO - Scottish Journal of Political Economy

    JF - Scottish Journal of Political Economy

    SN - 0036-9292

    ER -