The role of economic uncertainty in forecasting exchange rate returns and realized volatility: Evidence from quantile predictive regressions

• Author: Tahir Suleman,Christis Hassapis,Christina Christou,Rangan Gupta
• DOI: http://doi.org/10.1002/for.2539
• Published date: 01 November 2018
• Date: 01 November 2018

RESEARCH ARTICLE

The role of economic uncertainty in forecasting exchange

rate returns and realized volatility: Evidence from quantile

predictive regressions

Christina Christou

1

| Rangan Gupta

2

| Christis Hassapis

3

| Tahir Suleman

4,5

1

School of Economics, University of

Cyprus, Latsia, Cyprus

2

Department of Economics, University of

Pretoria, Gauteng, South Africa

3

Department of Economics, Faculty of

Social Sciences and Education, University

of Cyprus, Nicosia, Cyprus

4

Department of Economics and Finance,

Victoria Business School, Victoria

University of Wellington, Wellington,

New Zealand

5

School of Business, Wellington Institute

of Technology Lower Hutt, New Zealand

Correspondence

Tahir Suleman, Department of Economics

and Finance, Victoria Business School,

Victoria University of Wellington, Level 3,

Rutherford House, Pipitea Campus,

Wellington 6140, New Zealand.

Email: tahir.suleman@vuw.ac.nz

Abstract

In this paper, we investigate whether the news‐based measure of economic

policy uncertainty (EPU) can be used to forecast exchange rate returns and

volatility using a quantile regression approach, which accounts for persistence

and endogeneity, using data from 13 different countries. Our main findings

suggest that: (i) EPU is useful for forecasting exchange rate returns and

volatility, (ii) forecasting ability–quantile order relationships exhibit a U‐shape,

possibly asymmetric form around the median; and (iii) asymmetries are more

pronounced in the case of forecasting volatility.

KEYWORDS

developed and emerging markets, economic policy uncertainty, exchange rate returns, quantile

predictive regressions, volatility

1|INTRODUCTION

The foreign exchange market is the largest and most liq-

uid financial market in the world. As reported in the

Triennial Survey of global foreign exchange market

volumes, of the Bank for International Settlement (BIS),

the average daily turnover was 5.1 trillion US dollars in

April of 2016. In light of the importance of currency mar-

kets, accurate forecasting of exchange rate returns and

volatility is of paramount importance to various eco-

nomic agents. Forecasting of exchange rate is of interest,

not only to investors but also exporters and importers—

retailers and consumers—who ultimately take decisions

based on the value of the domestic currency and its

volatility. Moreover, policymakers are concerned with

pass‐through, a major mechanism by which the exchange

movements affect domestic economic aggregates. In this

regard, the literature on predictability of exchange rate

returns and volatility is voluminous, to say the least.

Detailed literature reviews are provided by Rapach and

Wohar (2002, 2004, 2006), Bacchetta and Van Wincoop

(2013), Rossi (2013), Plakandaras, Papadimitriou, and

Gogas (2013), Plakandaras, Papadimitriou, and Gogas

(2015), Plakandaras, Papadimitriou, Gogas, and

Konstantinos (2015), Plakandaras, Gupta, and Wohar

(2017), Pilbeam and Langeland (2015), Huber (2016,

2017), Papadimitriou, Gogas, and Plakandaras (2016),

Byrne, Korobilis, and Ribeiro (2016, 2018). One common

observation that emerges out of this literature is that,

despite the great need, the task of forecasting exchange rate

movements based on fundamentals is an arduous task.

In a recent study, Balcilar, Gupta, Kyei, and Wohar

(2016) provide evidence of in‐sample predictability of

returns and volatility, of 16 US dollar‐based exchange rates

Received: 27 November 2017 Revised: 11 June 2018 Accepted: 16 June 2018

DOI: 10.1002/for.2539

Journal of Forecasting. 2018;37:705–719. © 2018 John Wiley & Sons, Ltd.wileyonlinelibrary.com/journal/for 705