Roy KouwenbergAgnieszka MarkiewiczRalph VerhoeksRemco C.J. ZwinkelsMahidol UniversityErasmus School of EconomicsDe Nederlandsche BankVrije Universiteit Amsterdam2018-12-212019-03-142018-12-212019-03-142017-02-01Journal of Financial and Quantitative Analysis. Vol.52, No.1 (2017), 341-36317566916002210902-s2.0-85016047990https://repository.li.mahidol.ac.th/handle/20.500.14594/42139© 2017 Michael G. Foster School of Business, University of Washington. Exchange rate models with uncertain and incomplete information predict that investors focus on a small set of fundamentals that changes frequently over time. We design a model selection rule that captures the current set of fundamentals that best predicts the exchange rate. Out-of-sample tests show that the forecasts made by this rule significantly beat a random walk for 5 out of 10 currencies. Furthermore, the currency forecasts generate meaningful investment profits. We demonstrate that the strong performance of the model selection rule is driven by time-varying weights attached to a small set of fundamentals, in line with theory.Mahidol UniversityBusiness, Management and AccountingEconomics, Econometrics and FinanceReviewSCOPUS10.1017/S0022109017000011