Why is it so Difficult to Beat the Random Walk Rorecast of Exchange Rates?

By Lutz Kilian, Mark P. Taylor. Social Science Research Network, 2003

825 citations recorded by the source corpus.

We propose an exchange rate model that can explain both the observed volatility and the persistence of real and nominal exchange rate movements and thus in some measure resolves Rogoff’s (1996) purchasing power parity puzzle. Our analysis reconciles the well-known difficulties in beating the random walk forecast model with the statistical evidence of nonlinear mean reversion in deviations from fundamentals. Our analysis also provides a compelling rationale for the long-horizon predictability of exchange rates. We find strong empirical support for long-horizon predictability, and we explain why it is difficult to exploit this predictability in real-time forecasts. Our results not only lend support to economists’ beliefs that the exchange rate is inherently predictable, but they also help us to understand the reluctance of applied forecasters to abandon chartist methods in favor of models based on economic fundamentals.

Subfields: Economics and Econometrics

View Paper: https://www.semanticscholar.org/paper/9d2e7a9990db52caea61e15c7809adfba4d0a2c3

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Economics and Econometrics

Citations in the corpus

Asymptotic confidence intervals for impulse responses of near‐integrated processes (2004, background): “A natural extension of our approach is to explore the possibility that the mean reversion is non-linear due to the presence of transaction costs or noise trading and arbitrage (see Taylor 2001; Kilian and Taylor 2003; among others).”
Nonlinearity, Nonstationarity and Spurious Forecasts (2005, methodology): “The process modelled in this paper is oposite to that of Kilian and Taylor (2003); they describe nonlinear mean-reversion, however, combined with long-run predictability (they discuss forecasting of exchange rates).”
Real exchange rates and Purchasing Power Parity: mean-reversion in economic thought (2006, background): “A second potential source of nonlinearity was suggested by Kilian and Taylor (2003) as arising from the interaction of heterogeneous agents in the foreign exchange market at the microstructural level.”
Evaluating Direct Multistep Forecasts (2005, background): “Of course, a number of other studies, such as Marcellino ( 2002 ), Kilian and Taylor ( 2003 ), and Qi and Wu ( 2003 ), have considered DMS forecasts from nested nonlinear models.”
Real Exchange Rates Over the Past Two Centuries: How Important is the Harrod-Balassa-Samuelson Effect? (2008, positive): “The marginal signi&cance levels were calculated using the bootstrap algorithm described in Kilian and Taylor (2003).”
Can Oil Prices Forecast Exchange Rates (2011, neutral): “See also Faust, Rogers and Wright (2003), Kilian and Taylor (2003) and Engel, Mark and West (2007) for additional empirical evidence on predictive ability at longer horizons.”
Nonlinearities and Cyclical Behavior: The Role of Chartists and Fundamentalists (2003, background): “Our results throw doubt on the presumed stabilizing impact of fundamentalists on exchange rates, as proposed by Sarantis (1999) and Kilian and Taylor (2001).”
Testing for a unit root in a stationary ESTAR process (2011, methodology): “ESTAR models have been used to study the dynamics of real exchange rates (see Kilian and Taylor, 2003 ; Micheal et al, 1997 ; Taylor, 2001 ; Taylor et al, 2001 , among others).”
Concepts of equilibrium exchange rates (2005, methodology): “Even using non-linear models it is often difficult to beat a random walk except at long horizons; see, for example, Kilian and Taylor (2003).”
Tests for Unit-Root versus Threshold Specification With an Application to the Purchasing Power Parity Relationship (2004, background): “Following this idea, Michael, Nobay, and Peel (1997) and Kilian and Taylor (2003), among © 2004 American Statistical Association Journal of Business & Economic Statistics October 2004, Vol. 22, No. 4 DOI 10.1198/073500104000000389 382 others, estimated an exponential smooth transition…”
Forecast comparisons in unstable environments (2010, background): “Kilian and Taylor (2003) arrive at the same conclusion on the basis of an in-sample test that allows for nonlinearities in the data generating process (which is equivalent to parameter instabilities in predictive regressions).”
Nonparametric Bootstrap Procedures for Predictive Inference Based on Recursive Estimation Schemes (2007, methodology): “When we write the covariance kernel of the limiting distribution of the statistic (see below), however, we include terms capturing the contribution of parameter estimation error, thus implicitly assuming 9 See Hansen (2004) for further discussion of the White (2004) test. that g = qi for some i.”