Most TI discussion papers can be downloaded at

It is widely known that signiÞcant in-sample evidence of predictability does not guarantee signiÞcant out-of-sample predictability. This is often interpreted as an indication that in-sample evidence is likely to be spurious and should be discounted. In this paper we question this conventional wisdom. Our analysis shows that neither data mining nor parameter instability is a plausible explanation of the observed tendency of in-sample tests to reject the no predictability null more often than out-of-sample tests. We provide an alternative explanation based on the higher power of in-sample tests of predictability. We conclude that results of in-sample tests of predictability will typically be more credible than results of out-of-sample tests.

This paper provides a selective overview of puzzles in exchange rate economics. We begin with the forward bias puzzle: high interest rate currencies appreciate when one might guess that investors would demand higher interest rates on currencies expected to fall in value. We then analyze the purchasing power parity puzzle: the real exchange rate displays no (strong) reversion to a stable long-run equilibrium level. Finally, we cover the exchange rate disconnect puzzle: the lack of a link between the nominal exchange rate and economic fundamentals. For each puzzle, we critically review the literature and speculate on potential solutions. JEL classification: F31.

A major puzzle in international finance is the well-documented inability of models based on monetary fundamentals to produce better out-of-sample forecasts of the nominal exchange rate than a naive random walk. While this literature has generally employed statistical measures of forecast accuracy, we investigate whether there is any economic value to the predictive power of monetary fundamentals for the exchange rate. We find that, in the context of a simple asset allocation problem, the economic value of exchange rate forecasts from a fundamentals model can be greater than the economic value of random walk forecasts across a range of horizons.

This chapter discusses estimation, specification testing, and model selection of predictive density models. In particular, predictive density estimation is briefly discussed, and a variety of different specification and model evaluation tests due to various

the last decade or so, exchange rate economics has seen a number of important developments, with substantial contributions to both the theory and the empirical understanding of exchange rate determination. Important developments in econometrics and the increasing availability of high-quality data have also stimulated a large amount of empirical work on exchange rates. While this research has improved our understanding of exchange rates, a number of challenges and questions remain. One of the most widely studied and still unanswered questions in this literature involves why monetary models of exchange rate determination cannot forecast much of the variation in exchange rates. The monetary approach to exchange rate determination emerged as the dominant exchange rate model at the outset of the recent float in the early 1970s and remains an important exchange rate paradigm (Frenkel, 1976; Mussa, 1976, 1979; Bilson, 1978). However, Meese and Rogoff’s (1983a) finding that monetary models ’ forecasts could not outperform a simple no-change forecast was a devastating critique of standard models and marked a watershed in exchange rate economics. Moreover, even with the benefit of 20 years of hindsight, evidence that monetary models can consistently and significantly outperform a naïve random walk is still elusive (e.g., see Mark and Sul, 2001; Rapach and Wohar, 2001a, 2001b; Faust, Rogers, and Wright, 2001). This article reviews this puzzle and discusses several potential explanations for the consistent failure of monetary models to forecast much variation in nominal exchange rates. We present the essential elements of the monetary model in the

Using novel real-time data on a broad set of economic fundamentals for five major US dollar exchange rates over the recent float, we employ a predictive procedure that allows the relationship between exchange rates and fundamentals to evolve over time in a very general fashion. Our key findings are that: (i) the well-documented weak out-of-sample predictive ability of exchange rate models may be caused by poor performance of model-selection criteria, rather than lack of information content in the fundamentals; (ii) the difficulty of selecting the best predictive model is largely due to frequent shifts in the set of fundamentals driving exchange rates, reflecting swings in market expectations over time. However, the strength of the link between exchange rates and fundamentals is different across currencies.

In this paper a panel of vector error correction models based on a common long-run relationship is utilized to test whether the Euro exchange rates of Canada, Japan and the United States have a long-run link with monetary fundamentals. We use both exchange relationships relative to the full EMU area (with synthetic aggregates for the pre-EMU period) and relative to Germany solely. Compared to existing cointegration frameworks our approach provides more evidence that the aforementioned exchange rates are consistent with a rational expectations-based monetary exchange rate model based on a common long-run relationship, albeit with a longrun impact of relative income that is higher than predicted by the theory. As a next step we analyze the out-of-sample fit of this common long-run exchange rate model relative to naive random walk-based forecasts. These forecasting evaluations indicate that the monetary fundamentals-based common long-run model is superior to both random walk-based forecasts and standard cointegrated VAR model-based forecasts, especially at horizons of 2 to 4 years.

Since the breakdown of the Bretton Woods agreement, researchers have used a wide variety of structural models to try to predict exchange rate movements. Unfortunately, finding consistent evidence that these models outperform a random walk has proven elusive. In this paper we investigate the impact different methods of inference may have had on these conclusions. Using p-values based on recently developed tests of forecast accuracy and encompassing, as well as q-values designed to mitigate multiple testing problems, we provide stronger evidence consistent with these models having superior predictive ability. Our results suggest that previous studies ’ inability to detect predictive ability may have been influenced by the statistics used and the manner in which they were employed.

This paper proposes a simple chartist-fundamentalist model in which we allow for nonlinear time variation in chartists ’ extrapolation rate. Estimation of the model using monthly data for the major currencies vis-á-vis the US dollar shows that the model is significant in-sample and that it has out-of-sample predictive power for some of the currencies. We investigate the power of tests used in the literature to detect predictability against the alternative of the proposed model. We find that the short-term unpredictability and the long-term predictability are consistent with the model. The short-term unpredictability might be caused by the presence of weak nonlinearities that are difficult to detect at available sample sizes.