Structural Dynamic Factor Models (SDFM) represent a reliable tool to inform the construction of Dynamic Stochastic General Equilibrium (DSGE) models. The reason is that the log-linear solution of a DSGE model has a factor structure which ensures consistency between the representations of the two models. We assess the usefulness of SDFM for DSGE analysis by means of simulations. Using a standard DSGE model as the Data Generating Process, we show that the factor model always provides accurate estimates of the impulse response functions. As an application, we reassess the literature studying the response of hours to technology shock. An additional application studies the effects of monetary policy.
Forni, M., L., Gambetti, M., Lippi e L., Sala. "Informing DSGE Models through Dynamic Factor Models" Working paper, RECENT WORKING PAPER SERIES, Dipartimento di Economia Marco Biagi, 2025.
Informing DSGE Models through Dynamic Factor Models
Forni, M.;Gambetti, L.;Lippi, M.;
2025
Abstract
Structural Dynamic Factor Models (SDFM) represent a reliable tool to inform the construction of Dynamic Stochastic General Equilibrium (DSGE) models. The reason is that the log-linear solution of a DSGE model has a factor structure which ensures consistency between the representations of the two models. We assess the usefulness of SDFM for DSGE analysis by means of simulations. Using a standard DSGE model as the Data Generating Process, we show that the factor model always provides accurate estimates of the impulse response functions. As an application, we reassess the literature studying the response of hours to technology shock. An additional application studies the effects of monetary policy.File | Dimensione | Formato | |
---|---|---|---|
REcent-wp160.pdf
Open access
Tipologia:
VOR - Versione pubblicata dall'editore
Dimensione
1.96 MB
Formato
Adobe PDF
|
1.96 MB | Adobe PDF | Visualizza/Apri |
Pubblicazioni consigliate
I metadati presenti in IRIS UNIMORE sono rilasciati con licenza Creative Commons CC0 1.0 Universal, mentre i file delle pubblicazioni sono rilasciati con licenza Attribuzione 4.0 Internazionale (CC BY 4.0), salvo diversa indicazione.
In caso di violazione di copyright, contattare Supporto Iris