Narrow Search
Last searches

Results for *

Displaying results 1 to 2 of 2.

  1. Predicting fiscal crises
    a machine learning approach
    Published: May 2021
    Publisher:  International Monetary Fund, [Washington, D.C.]

    In this paper I assess the ability of econometric and machine learning techniques to predict fiscal crises out of sample. I show that the econometric approaches used in many policy applications cannot outperform a simple heuristic rule of thumb.... more

    Access:
    Verlag (kostenfrei)
    Verlag (kostenfrei)
    Resolving-System (kostenfrei)
    Staatsbibliothek zu Berlin - Preußischer Kulturbesitz, Haus Unter den Linden
    Unlimited inter-library loan, copies and loan

     

    In this paper I assess the ability of econometric and machine learning techniques to predict fiscal crises out of sample. I show that the econometric approaches used in many policy applications cannot outperform a simple heuristic rule of thumb. Machine learning techniques (elastic net, random forest, gradient boosted trees) deliver significant improvements in accuracy. Performance of machine learning techniques improves further, particularly for developing countries, when I expand the set of potential predictors and make use of algorithmic selection techniques instead of relying on a small set of variables deemed important by the literature. There is considerable agreement across learning algorithms in the set of selected predictors: Results confirm the importance of external sector stock and flow variables found in the literature but also point to demographics and the quality of governance as important predictors of fiscal crises. Fiscal variables appear to have less predictive value, and public debt matters only to the extent that it is owed to external creditors

     

    Export to reference management software   RIS file
      BibTeX file
    Source: Staatsbibliothek zu Berlin
    Language: English
    Media type: Ebook
    Format: Online
    ISBN: 9781513573588
    Other identifier:
    Series: IMF working paper ; WP/21, 150
    Subjects: Early warning systems; sovereign default; random forest; Foreign Exchange; Informal Economy; Underground Econom
    Scope: 1 Online-Ressource (circa 66 Seiten), Illustrationen
  2. Predicting fiscal crises
    a machine learning approach
    Published: May 2021
    Publisher:  International Monetary Fund, [Washington, D.C.]

    In this paper I assess the ability of econometric and machine learning techniques to predict fiscal crises out of sample. I show that the econometric approaches used in many policy applications cannot outperform a simple heuristic rule of thumb.... more

    Access:
    Verlag (kostenfrei)
    Verlag (kostenfrei)
    Resolving-System (kostenfrei)
    Verlag (kostenfrei)
    Orient-Institut Beirut
    Online
    No inter-library loan
    Staatsbibliothek zu Berlin - Preußischer Kulturbesitz, Haus Potsdamer Straße
    No inter-library loan
    Universitätsbibliothek Braunschweig
    No inter-library loan
    Staats- und Universitätsbibliothek Bremen
    No inter-library loan
    Universitätsbibliothek Erfurt / Forschungsbibliothek Gotha, Universitätsbibliothek Erfurt
    No inter-library loan
    Bibliothek der Pädagogischen Hochschule Freiburg/Breisgau
    No inter-library loan
    Niedersächsische Staats- und Universitätsbibliothek Göttingen
    No inter-library loan
    Universitäts- und Landesbibliothek Sachsen-Anhalt / Zentrale
    No inter-library loan
    Helmut-Schmidt-Universität, Universität der Bundeswehr Hamburg, Universitätsbibliothek
    No inter-library loan
    Technische Universität Hamburg, Universitätsbibliothek
    No inter-library loan
    Technische Informationsbibliothek (TIB) / Leibniz-Informationszentrum Technik und Naturwissenschaften und Universitätsbibliothek
    No inter-library loan
    Duale Hochschule Baden-Württemberg Heidenheim, Bibliothek
    e-Book Nationallizenz
    No inter-library loan
    Thüringer Universitäts- und Landesbibliothek
    No inter-library loan
    Fachhochschule Kiel, Zentralbibliothek
    No inter-library loan
    ZBW - Leibniz-Informationszentrum Wirtschaft, Standort Kiel
    VS 301
    No inter-library loan
    Universitätsbibliothek Leipzig
    No inter-library loan
    Leuphana Universität Lüneburg, Medien- und Informationszentrum, Universitätsbibliothek
    No inter-library loan
    Duale Hochschule Baden-Württemberg Mosbach, Bibliothek
    E-Book Nationallizenz IMF
    No inter-library loan
    Hochschule Offenburg, University of Applied Sciences, Bibliothek Campus Offenburg
    E-Book International Monetary Fund
    No inter-library loan
    Hochschulbibliothek Pforzheim, Bereichsbibliothek Technik und Wirtschaft
    e-Book International Monetary Fund eLibrary
    No loan of volumes, only paper copies will be sent
    Duale Hochschule Baden-Württemberg Ravensburg, Bibliothek
    E-Book IMF
    No inter-library loan
    Hochschule Albstadt-Sigmaringen, Bibliothek Sigmaringen
    No loan of volumes, only paper copies will be sent
    Duale Hochschule Baden-Württemberg Villingen-Schwenningen, Bibliothek
    E_Book IMF
    No inter-library loan

     

    In this paper I assess the ability of econometric and machine learning techniques to predict fiscal crises out of sample. I show that the econometric approaches used in many policy applications cannot outperform a simple heuristic rule of thumb. Machine learning techniques (elastic net, random forest, gradient boosted trees) deliver significant improvements in accuracy. Performance of machine learning techniques improves further, particularly for developing countries, when I expand the set of potential predictors and make use of algorithmic selection techniques instead of relying on a small set of variables deemed important by the literature. There is considerable agreement across learning algorithms in the set of selected predictors: Results confirm the importance of external sector stock and flow variables found in the literature but also point to demographics and the quality of governance as important predictors of fiscal crises. Fiscal variables appear to have less predictive value, and public debt matters only to the extent that it is owed to external creditors

     

    Export to reference management software   RIS file
      BibTeX file
    Source: Staatsbibliothek zu Berlin
    Language: English
    Media type: Ebook
    Format: Online
    ISBN: 9781513573588
    Other identifier:
    Series: IMF working paper ; WP/21, 150
    Subjects: Early warning systems; sovereign default; random forest; Foreign Exchange; Informal Economy; Underground Econom; Macroeconomics; Economics: General; International Economics; Economic & financial crises & disasters; Economics of specific sectors; Currency crises; Informal sector; Economics
    Scope: 1 Online-Ressource (circa 66 Seiten), Illustrationen