Essays in Applied Macroeconometrics

Moramarco, Graziano (2019) Essays in Applied Macroeconometrics, [Dissertation thesis], Alma Mater Studiorum Università di Bologna. Dottorato di ricerca in Economics, 31 Ciclo. DOI 10.6092/unibo/amsdottorato/8956.
Documenti full-text disponibili:
[img] Documento PDF (English) - Richiede un lettore di PDF come Xpdf o Adobe Acrobat Reader
Disponibile con Licenza: Salvo eventuali più ampie autorizzazioni dell'autore, la tesi può essere liberamente consultata e può essere effettuato il salvataggio e la stampa di una copia per fini strettamente personali di studio, di ricerca e di insegnamento, con espresso divieto di qualunque utilizzo direttamente o indirettamente commerciale. Ogni altro diritto sul materiale è riservato.
Download (3MB)


This thesis consists of three papers. The first paper provides new measures of global and country-specific macroeconomic uncertainty using a global vector autoregressive (GVAR) model. The approach accounts for the international propagation of uncertainty, captures global uncertainty shocks and quantifies cross-country uncertainty spillovers. Due to global economic linkages, measured uncertainty is highly correlated across countries. Moreover, the paper exploits the error correction representation of the GVAR to distinguish between short-run and long-run uncertainty, and shows that such distinction may help reconcile popular indicators of uncertainty, such as the VIX and the economic policy uncertainty index by Baker, Bloom and Davis (2016). The second paper documents the predictive potential of financial-cycle indicators for real economic activity through an extensive comparative evaluation using a large dataset for the United States. The study examines in-sample and out-of-sample performance and explores a variety of methods for data-rich environments. A cyclically-adjusted house price-rent ratio and the liabilities-income ratio of the non-corporate business sector appear as the most powerful predictors of GDP over horizons between 1 and 7 years in recent decades. Overall, financial-cycle variables are shown to provide valuable predictive content both individually and collectively. High-dimensional models and forecast combinations using all available data fail to outperform the best small models. The third paper investigates an approach for enhancing density forecasts of macroeconomic variables using Bayesian Markov-switching models. Alternative views on economic regimes are pooled to form flexible composite forecasts, which are optimized with respect to log scores and probability integral transforms. In an application to U.S. GDP, the approach produces well-calibrated forecasts, achieves better calibration than competing methods and delivers good accuracy in terms of predictive densities. The proposed procedure also evaluates the time-varying contribution of different views to calibration and accuracy. The empirical application examines views derived from the Fed supervisory scenarios.

Tipologia del documento
Tesi di dottorato
Moramarco, Graziano
Dottorato di ricerca
Settore disciplinare
Settore concorsuale
Parole chiave
Global uncertainty; GVAR; uncertainty indices; uncertainty spillovers; financial cycle; business cycle; GDP forecasts; housing; macro-finance; high-dimensional models; density forecasts; Bayesian; Markov-switching; forecast combination
Data di discussione
19 Marzo 2019

Altri metadati

Statistica sui download

Gestione del documento: Visualizza la tesi