An In-Sample Evaluation of Exchange Rate Models: In Search of Scapegoats

Yin-Wong Cheung, Wenhao Wang and Frank Westermann

No 11852, CESifo Working Paper Series from CESifo

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.

Keywords: Bayesian dynamic model averaging; explaining exchange rates; in-sample performance; purchasing power parity deviations. (search for similar items in EconPapers)
JEL-codes: C11 F31 (search for similar items in EconPapers)
Date: 2025
New Economics Papers: this item is included in nep-opm
References: View references in EconPapers View complete reference list from CitEc
Citations:

Downloads: (external link)
https://www.ifo.de/DocDL/cesifo1_wp11852.pdf (application/pdf)

Related works:
Working Paper: An In-Sample Evaluation of Exchange Rate Models: In Search of Scapegoats (2025)
This item may be available elsewhere in EconPapers: Search for items with the same title.

Export reference: BibTeX RIS (EndNote, ProCite, RefMan) HTML/Text

Persistent link: https://EconPapers.repec.org/RePEc:ces:ceswps:_11852

Access Statistics for this paper

More papers in CESifo Working Paper Series from CESifo Contact information at EDIRC.
Bibliographic data for series maintained by Klaus Wohlrabe ().