A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation

This study revisits the Meese-Rogoff puzzle by estimating the traditional monetary models of exchange rate determination in state-space form and comparing the accuracy of these forecasts against the naïve random walk model using a wide range of conventional and alternative measures of forecasting ac...

Full description

Bibliographic Details
Main Author: Burns, Kelly
Format: Journal Article
Published: Multinational Finance Society 2016
Online Access:http://hdl.handle.net/20.500.11937/74380
_version_ 1848763258756923392
author Burns, Kelly
author_facet Burns, Kelly
author_sort Burns, Kelly
building Curtin Institutional Repository
collection Online Access
description This study revisits the Meese-Rogoff puzzle by estimating the traditional monetary models of exchange rate determination in state-space form and comparing the accuracy of these forecasts against the naïve random walk model using a wide range of conventional and alternative measures of forecasting accuracy. The results demonstrate that incorporating stochastic movements in the parameters of exchange rate models does not enable the Meese-Rogoff puzzle to be overturned. However, estimating these models in state-space form substantially improves forecasting accuracy to the extent that the model and random walk produce an equivalent magnitude of error. Furthermore, the results prove that the Meese-Rogoff puzzle can be overturned if the forecasts are evaluated by alternative criteria. These criteria include direction accuracy, profitability, and measures that jointly take into account both magnitude and direction accuracy.
first_indexed 2025-11-14T11:00:36Z
format Journal Article
id curtin-20.500.11937-74380
institution Curtin University Malaysia
institution_category Local University
last_indexed 2025-11-14T11:00:36Z
publishDate 2016
publisher Multinational Finance Society
recordtype eprints
repository_type Digital Repository
spelling curtin-20.500.11937-743802019-02-19T04:16:42Z A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation Burns, Kelly This study revisits the Meese-Rogoff puzzle by estimating the traditional monetary models of exchange rate determination in state-space form and comparing the accuracy of these forecasts against the naïve random walk model using a wide range of conventional and alternative measures of forecasting accuracy. The results demonstrate that incorporating stochastic movements in the parameters of exchange rate models does not enable the Meese-Rogoff puzzle to be overturned. However, estimating these models in state-space form substantially improves forecasting accuracy to the extent that the model and random walk produce an equivalent magnitude of error. Furthermore, the results prove that the Meese-Rogoff puzzle can be overturned if the forecasts are evaluated by alternative criteria. These criteria include direction accuracy, profitability, and measures that jointly take into account both magnitude and direction accuracy. 2016 Journal Article http://hdl.handle.net/20.500.11937/74380 Multinational Finance Society restricted
spellingShingle Burns, Kelly
A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation
title A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation
title_full A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation
title_fullStr A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation
title_full_unstemmed A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation
title_short A Reconsideration of the Meese-Rogoff Puzzle – Alternative Approaches to Model Estimation and Forecast Evaluation
title_sort reconsideration of the meese-rogoff puzzle – alternative approaches to model estimation and forecast evaluation
url http://hdl.handle.net/20.500.11937/74380