• Research Support
      • Starting Library Research
      • Research Guides
      • Subject Librarians
      • Digital Scholarship Center
      • Publishing & Copyright
      • Research Data Management
      • Study Skills & Campus Resources
      • Request a Purchase
    • Using the Libraries
      • Rooms & Study Spaces
      • Borrowing & Requesting
      • Connect from Off-campus
      • ILL (Interlibrary Loan)
      • Course Reserves
      • Technology, Printing, & Scanning
      • Tutoring at the UO Libraries
      • Classroom Technology Support
      • Canvas Support
      • Accessibility
    • Collections
      • Databases A-Z
      • Scholars' Bank
      • Videos, Music, Photos
      • Special Collections & University Archives
      • Unique Collections
      • Government Documents
      • Maps & Aerial Photography
      • Oregon Digital
      • Oregon Newspapers
    • Library Accounts
      • LibrarySearch Account
      • ILLiad Account (Interlibrary Loan)
      • Endnote Web (Citation Manager)
    • About
      • Hours & Locations
      • Staff & Department Directory
      • News & Events
      • Calendar
      • Mission, Values, & Strategic Directions
      • Diversity & Inclusion
      • Policies
      • Jobs
      • Comments & Suggestions
    • Chat/Ask Us
      • Email
      • Phone
      • Text
    View Item 
    •   Scholars' Bank Home
    • University Archives
    • University of Oregon Administration
    • The Office of Research, Innovation and Graduate Education
    • Theses and Dissertations
    • View Item
    •   Scholars' Bank Home
    • University Archives
    • University of Oregon Administration
    • The Office of Research, Innovation and Graduate Education
    • Theses and Dissertations
    • View Item
    JavaScript is disabled for your browser. Some features of this site may not work without it.

    Incorporating High Dimensional Data Vectors into Structural Macroeconomic Models

    View/Open
    Gelfer_oregon_0171A_11579.pdf (14.32Mb)

    Date
    2016-10-27
    Author
    Gelfer, Sacha
    Metadata
    Show full item record
    Author
    Gelfer, Sacha
    Abstract
    In this dissertation I incorporate high dimensional data vectors in estimated Dynamic Stochastic General Equilibrium (DSGE) models, evaluating the labor market dynamics incorporated inside such data vectors, out-of-sample forecasting performance of many models estimated with such data vectors and analytically examining the reduction of macroeconomic volatility that can occur when such data vectors are used in the formation of expectations about the future. The second chapter investigates the extent to which modern DSGE models can produce labor market dynamics in response to a financial crisis that are consistent with the experience of the Great Recession. I estimate two New-Keynesian models, one with and one without financial frictions, in a data-rich environment. I find that negative financial shocks are associated with longer recoveries in real investment, capital-intensive sectors of the labor market and average unemployment duration. I also find the model with a financial accelerator is equipped with better tools to identify the dynamics associated with the Great Recession and its recovery in regard to many labor and financial metrics. The third chapter compares the out-of-sample forecasting performance of the two DSGE models of Chapter II when they are estimated both out of and in a data-rich environment. This chapter finds that many financial time series variance decomposition are significantly better explained using the structural set-up of the New-Keynesian model with financial frictions. DSGE models estimated with high dimensional data vectors significantly out forecast their regularly estimated counterpart in regard to output, investment and consumption growth. Lastly, the use of real-time optimal pool model weighting significantly out-forecasts traditional macroeconomic models as well as an equally weighted weighting scheme in terms of many macroeconomic variables. The fourth chapter examines the role forecasts derived by high dimensional data vectors can have on lowering macroeconomic volatility. Bounded rational agents are introduced into the Chapter II DSGE model with financial frictions and are given the option to use or ignore professionally generated forecasts from a dynamic factor model in their perceived forecasting model. In simulations, I find that professionally generated forecasts can significantly lower the volatility of many macroeconomic variables including inflation and hours worked.
    URI


    Collections
    • Theses and Dissertations [2826]
    • Economics Theses and Dissertations [61]

    My Account

    LoginRegister

    Browse

    All of Scholars' BankCommunities & CollectionsBy Issue DateAuthorsTitlesSubjectsThis CollectionBy Issue DateAuthorsTitlesSubjects

    Statistics

    View Usage Statistics

    CAMPUS

    • News
    • Events
    • Maps
    • Directions
    • Class Schedule
    • Academic Calendar

    RESOURCES

    • Webmail
    • Canvas
    • Duckweb
    • Find People
    • A–Z Index
    • UO Libraries

    RELATED INFORMATION

    • Library Accessibility
    • Privacy Policy
    • Giving
    UO Libraries
    1501 Kincaid Street
    1239 University of Oregon
    Eugene, OR 97403-1299

    T: 541-346-3053

    FacebookTwitterYouTubeInstagram

    CareersPrivacy PolicyAboutFind People
    University of Oregon
    ©University of Oregon. All Rights Reserved.

    UO prohibits discrimination on the basis of race, color, sex, national or ethnic origin, age, religion, marital status, disability, veteran status, sexual orientation, gender identity, and gender expression in all programs, activities and employment practices as required by Title IX, other applicable laws, and policies. Retaliation is prohibited by UO policy. Questions may be referred to the Title IX Coordinator, Office of Affirmative Action and Equal Opportunity, or to the Office for Civil Rights. Contact information, related policies, and complaint procedures are listed on the statement of non-discrimination.