BIROn - Birkbeck Institutional Research Online

    Bayesian Policy learning during COVID-19

    Mamatzakis, Emmanuel and Ongena, S. and Tsionas, M. (2021) Bayesian Policy learning during COVID-19. Working Paper. Birkbeck, University of London, London, UK. (Unpublished)

    [img]
    Preview
    Text
    paper_COVID 15July21.pdf - Published Version of Record

    Download (1MB) | Preview

    Abstract

    The rapid spread of COVID-19 across the globe primed a variety of non-pharmaceutical interventions (NPIs). Given these NPIs, whether the SIR parameters followed a Bayesian learning, a random walk pattern or other type of learning with evolving epidemiological data over time has implications for policy learning literature. Using a sample of UK country specific data and also for 168 countries and 51,083 country-date observations (January 1, 2020 to January 9, 2021), we estimate a SIR model with time-varying β and γ parameters in three context of a dynamic panel vector autoregressive model. Although learning does not seem to be taking place, and despite the absence of evidence of governments’ learning from the past, most policy measures are effective in reducing the values of the β and γ parameters. We also provide estimates of time-varying β and γ that can be used widely, and we develop novel testing procedures for testing for Bayesian learning.

    Metadata

    Item Type: Monograph (Working Paper)
    School: School of Business, Economics & Informatics > Management
    Research Centres and Institutes: Accounting and Finance Research Centre
    Depositing User: Emmanuel Mamatzakis
    Date Deposited: 20 Jul 2021 05:48
    Last Modified: 24 Jul 2021 07:18
    URI: https://eprints.bbk.ac.uk/id/eprint/45203

    Statistics

    Downloads
    Activity Overview
    14Downloads
    35Hits

    Additional statistics are available via IRStats2.

    Archive Staff Only (login required)

    Edit/View Item Edit/View Item