On the Predictive Information of Futures' Prices: A Wavelet‐Based Assessment

• Author: Stephan Schlüter,Helmut Herwartz
• Date: 01 July 2017
• Published date: 01 July 2017
• DOI: http://doi.org/10.1002/for.2435

Journal of Forecasting,J. Forecast. 36, 345–356 (2017)

Published online 26 August 2016 in Wiley Online Library (wileyonlinelibrary.com)DOI: 10.1002/for.2435

On the Predictive Information of Futures’ Prices: A

Wavelet-Based Assessment

HELMUT HERWARTZ1AND STEPHAN SCHLÜTER2

1

Department of Statistics and Econometrics, University of Göttingen, Germany

2

Department of Mathematics, Natural & Economic Science, University of Applied Sciences

Ulm, Germany

ABSTRACT

While in speculative markets forward prices could be regarded as natural predictors for future spot rates, empirically,

forward prices often fail to indicate ex ante the direction of price movements. In terms of forecasting, the random walk

approximation of speculative prices has been established to provide ‘naive’ predictors that are most difficult to out-

perform by both purely backward-looking time series models and more structural approaches processing information

from forward markets. We empirically assess the implicit predictive content of forward prices by means of wavelet-

based prediction of two foreign exchange (FX) rates and the price of Brent oil quoted either in US dollars or euros.

Essentially, wavelet-based predictors are smoothed auxiliary (padded) time series quotes that are added to the sample

information beyond the forecast origin. We compare wavelet predictors obtained from padding with constant prices

(i.e. random walk predictors) and forward prices. For the case of FX markets, padding with forward prices is more

effective than padding with constant prices, and, moreover, respective wavelet-based predictors outperform purely

backward-looking time series approaches (ARIMA). For the case of Brent oil quoted in US dollars, wavelet-based

predictors do not signal predictive content of forward prices for future spot prices. Copyright © 2016 John Wiley &

Sons, Ltd..

KEY WORDS forecasting; futures prices; wavelet transform; denoising

INTRODUCTION

In the context of ex ante forecasting spot prices in speculative markets the predictive content of forward market

prices (futures’ prices) has been widely and controversially discussed (Fama, 1984; Hodrick, 1989; Engel, 1996).

Economically, term (or risk) premia are natural candidates to explain systematic deviations between current forward

prices and future spot prices. With regard to foreign exchange (FX) markets, for instance, the uncovered interest rate

parity establishes that forward FX rates should convey information on future spot rates. In the empirical literature on

FX rate forecasting, however, the so-called forward rate bias puzzle has attracted considerable attention. Numerous

empirical studies have questioned the predictive information of forward FX rates, or motivate why forward rates may

systematically fail to signal ex ante the direction of spot price changes (Froot and Thaler, 1990; Chinn and Meredith,

2004; Chakraborty and Evans, 2008; Burnside et al., 2011, Menkhoff et al., 2012). Exemplifying a second set of

markets, forward commodity prices might comprise specific mark-ups (so-called convenience yields) that relate to

the physical character of the traded assets. Examples for such ‘premia’ might include cost of transport and/or storage,

with the latter being of particular relevance under scarce capacity.Moreover, future spot prices on commodity markets

are often timely reflectors of political uncertainty or crises (and wars) that appear hardly predictable in general. For

both reasons, forward commodity prices might lack signaling value for future spot quotes. Against a background of

distinguished market characteristics (Routhledge et al., 2000), the predictive content of forward commodity/oil prices

has also seen a controversial discussion (Casassus and Collin-Dufresne, 2005; Alquist et al., 2011). With regard to

oil markets as very prominent representatives of commodity exchanges, Reef and Vigfusson (2011) and, similarly,

Yanagisawa (2009) argue in favor of predictive content of forward prices on oil markets. On the downside, however,

there is ample work questioning the ex ante predictive content of oil futures (Alquist and Kilian, 2010; Knetsch, 2007;

Baumeister and Kilian, 2011).

On the statistical/econometric side, tests of forward premium puzzles have benefited from modern approaches to

persistency measurement. Maynard (2006) argues that evidence against unbiasedness of forward rates is weakened

when employing regression diagnostics that are robust to persistent regressors. Similarly, persistence differentials

among spot returns and forward premia have been suggested to explain the failure of futures’ prices to predict future

spot prices (Maynard and Phillips, 2001). In light of the recent statistical/econometric literature, it appears that the for-

ward bias puzzle can be considered a statistical artifact rather than an economic puzzle. Relating to modern concepts

Correspondence to: Stephan Schlüter, Wingas GmbH, Kassel, Germany. E-mail: stephan.schlueter@hs-ulm.de

Copyright © 2016 John Wiley & Sons, Ltd.