Forecasting the volatility of Nikkei 225 futures

• DOI: http://doi.org/10.1002/fut.21847
• Published date: 01 November 2017
• Date: 01 November 2017

Received: 7 August 2015

|

Accepted: 2 February 2017

DOI: 10.1002/fut.21847

RESEARCH ARTICLE

Forecasting the volatility of Nikkei 225 futures

Manabu Asai

1

|

Michael McAleer

2,3,4,5

1

Faculty of Economics, Soka University,

Tokyo, Japan

2

Department of Quantitative Finance,

National Tsing Hua University, Taiwan,

China

3

Department of Econometrics Netherlands,

Erasmus Universiteit Rotterdam, Rotterdam,

The Netherlands

4

Department of Quantitative Economics,

Complutense University of Madrid, Madrid,

Spain

5

Institute of Advanced Sciences,

Yokohama National University, Yokohama,

Japan

Correspondence

Manabu Asai, Faculty of Economics, Soka

University, 1-236 Tangi-cho, Hachioji,

Tokyo 192-8577, Japan.

Email: m-asai@soka.ac.jp

Funding information

Japan Society for the Promotion of Science,

Grant number: JP16K03603

This article proposes an indirect method for forecasting the volatility of futures

returns, based on the relationship between futures and the underlying asset for the

returns and time-varying volatility. The paper considers the stochastic volatility

model with asymmetry and long memory, using high frequency data of the underlying

asset, for forecasting its volatility. The empirical results for Nikkei 225 futures

indicate that the adjusted R

2

supports the appropriateness of the indirect method, and

that the new method based on stochastic volatility models with asymmetry and long

memory outperforms the forecasting model based on the direct method using the

pseudo long time series.

JEL CLASSIFICATION

C22, C53, C58, G17

1

|

INTRODUCTION

It is essential to forecast financial market volatility in both the short and long run for optimal risk management and hedging

strategies. For example, under the Basel II Accord, banks and other authorized deposit-taking institutions need to use short-term

volatility forecasts to produce daily Value-at-Risk (VaR) measures, whereas they use longer term volatility forecasts for option

pricing and asset allocation. Consequently, the past two decades have witnessed a growing literature on forecasting volatility of

returns on futures contracts (see, e.g., Simon, 2002; Hong, Nohel, & Todd, 2015).

In termsof time series forecasting, the availablesample size before each maturitydate is generally insufficientto use time series

models, suchas autoregressive moving-average (ARMA),autoregressive fractionally integratedmoving-average (ARFIMA),and

generalizedautoregressive conditional heteroskedasticity (GARCH)models. In order to alleviate this problem,the datasets used in

Jorion (1995), Martens (2002), Sadorsky (2006), Lai and Sheu (2010), and Lai (2016), among others, are based on the prices of

futures contracts closest to maturity, in order to connect small-sized datasets to create a pseudo long time series.

Manabu Asai is a Professor of Econometrics, Faculty of Economics, Soka University, Tokyo, Japan. Michael McAleer is a University Distinguished Chair

Professor, Department of Quantitative Finance, National Tsing Hua University, Taiwan, a Professor of Quantitative Finance, Erasmus Universiteit

Rotterdam, Department of Econometrics Netherlands, Rotterdam, a Distinguished Visiting Professor, Department of Quantitative Economics, Complutense

University of Madrid, Spain, and a IAS Adjunct Professor, Institute of Advanced Sciences, Yokohama National University, Japan.

J Futures Markets. 2017;37:1141–1152. wileyonlinelibrary.com/journal/fut © 2017 Wiley Periodicals, Inc.

|

1141