Please use this identifier to cite or link to this item: https://hdl.handle.net/10419/338309 
Year of Publication: 
2025
Series/Report no.: 
CESifo Working Paper No. 11852
Version Description: 
This Version: November 2025
Publisher: 
Munich Society for the Promotion of Economic Research - CESifo GmbH, Munich
Abstract: 
We employ a modified dynamic model averaging framework which permits inferences about the shifting relevance and significance of explanatory variables to assess in-sample performance of exchange-rate models and empirical validity of purchasing-power-parity (PPP). The analysis is based on 16,384 empirical specifications constructed from 14 canonical and newly introduced explanatory variables for six US dollar exchange rates. Our findings indicate the best performing empirical exchange rate specification is unstable and changes frequently; individual explanatory variables display large time- and cross-currency variation in relevance and effects; and the combination of explanatory variables that enhances the empirical evidence of PPP differs by exchange rates. These findings underscore the challenge in applying a single exchange-rate model or the scapegoat hypothesis to explain all exchange rates in all historical periods.
Subjects: 
Bayesian Dynamic Model Averaging
Explaining Exchange Rates
In-Sample Performance
Purchasing Power Parity Deviations
JEL: 
C11
F31
Document Type: 
Working Paper
Appears in Collections:

Files in This Item:
File
Size





Items in EconStor are protected by copyright, with all rights reserved, unless otherwise indicated.