The dynamic correlations between the G7 economies and China: Evidence from both realized and implied volatilities

• Date: 01 October 2017
• Published date: 01 October 2017
• Author: Xuyuanda Qi,Xingguo Luo
• DOI: http://doi.org/10.1002/fut.21851

Received: 28 February 2017

|

Accepted: 28 February 2017

DOI: 10.1002/fut.21851

RESEARCH ARTICLE

The dynamic correlations between the G7 economies and China:

Evidence from both realized and implied volatilities

Xingguo Luo

1

|

Xuyuanda Qi

2

1

College of Economics and Academy of

Financial Research, Zhejiang University,

Hangzhou, Zhejiang, China

2

Simon Business School, University of

Rochester, Rochester, New York

Correspondence

Xingguo Luo, College of Economics and

Academy of Financial Research, Zhejiang

University, 38 Zheda RD, Hangzhou,

Zhejiang, China.

Email: xingguoluo@gmail.com

Funding information

National Natural Science Foundation of

China, Grant number: 71301143; Natural

Science Foundation of Zhejiang Province,

Grant number: LQ13G030001

This paper investigates the dynamic correlations between the G7 economies and

China by using the EGARCH/DCC models proposed by Engle and Figlewski (2015).

We find that the correlations between the G7 economies can be captured by a one-

factor model when either the realized or implied volatilities are used. Although no

significant correlations between China and the G7 countries are captured using

realized volatilities, we find that the correlations increased during the 2008 financial

crisis. Furthermore, we show that the one-factor model is useful for hedging the

volatility risks of individual countries.

JEL CLASSIFICATION

C32, G12, G13

1

|

INTRODUCTION

Engle and Figlewski (2015) develop EGARCH/DCC models to document a general correlation structure and a one-factor model,

with the VIX as the common factor among the implied volatilities (IV) of 28 large cap stocks. Their empirical study shows that

IV changes are correlated through exposure to a single market volatility process, that is, the VIX index. Furthermore, Engle and

Figlewski demonstrate that their findings can help us to build up a dynamic hedge for the vega risk with a hedging portfolio of IV

exposures. Krause and Lien (2014) extend this model, and consider both the market volatility factor and the industry volatility

factor in the evolution of individual stock option implied volatility.

A number of researchers such as Ball and Torous (2000) and Buraschi, Porchia, and Trojani (2010) emphasize the

importance of correlations across countries for making international portfolio choices. In particular, Bekaert, Ehrmann,

Fratzscher, and Mehl (2014) find evidence of contagion in the global markets. However, these studies consider return

correlations, and it is unclear what correlation structure exists among the volatilities of different countries. A new study by Yang

and Zhou (2017) examines volatility spillovers across countries in terms of implied volatility. Moreover, given that the

difference between realized and implied volatility is highly debated by researchers such as Canina and Figlewski (1993),

Christensen and Prabhala (1998), Becker, Clements, and White (2007) and Kourtis, Markellos, and Symeonidis (2016), it is

important to consider realized and implied volatility separately.

We are grateful for helpful comments from Robert Engle, Bob Webb (the editor), Gary Yang (our VINS conference discussant), Yang Zhao (our ICFDM

discussant), and participants at the First Annual Volatility Institute at NYU Shanghai (VINS) Conference in Shanghai and the Fifth International

Conference on Futures and Derivative Markets (ICFDM) in Shenzhen. Xingguo Luo has been supported by the National Natural Science Foundation of

China (project no. 71301143) and the Natural Science Foundation of Zhejiang Province (project no. LQ13G030001).

J Futures Markets. 2017;37:989–1002. wileyonlinelibrary.com/journal/fut © 2017 Wiley Periodicals, Inc.

|

989