Forecasting stock volatility in the presence of extreme shocks: Short‐term and long‐term effects

• Published date: 01 August 2020
• Date: 01 August 2020
• Author: Guoshan Liu,Lu Wang,Feng Ma
• DOI: http://doi.org/10.1002/for.2668

RESEARCH ARTICLE

Forecasting stock volatility in the presence of extreme

shocks: Short-term and long-term effects

Lu Wang

1

| Feng Ma

2

| Guoshan Liu

1

1

School of Mathematics, Southwest

Jiaotong University, Chengdu, China

2

School of Economics and Management,

Southwest Jiaotong University, Chengdu,

China

Correspondence

Feng Ma, School of Economics and

Management, Southwest Jiaotong

University, No. 111, North 1st Section,

2nd Ring Road, Chengdu, 610031, China.

Email: mafeng2016@swjtu.edu.cn

Funding information

Double First-Class project of Southwest

Jiaotong University, Grant/Award

Number: JDSYLYB2018033; Key

Laboratory of Statistical information

technology and data mining, State

Statistics Bureau, Grant/Award Number:

SDL201711; Humanities and Social

Science Fund of the Ministry of Education,

Grant/Award Numbers: 17YJC790105,

17XJCZH002; National Natural Science

Foundation of PR China, Grant/Award

Numbers: 71671145, 71701170

Abstract

This paper introduces a novel generalized autoregressive conditional

heteroskedasticity–mixed data sampling–extreme shocks (GARCH-MIDAS-ES)

model for stock volatility to examine whether the importance of extreme

shocks changes in different time ranges. Based on different combinations of

the short- and long-term effects caused by extreme events, we extend the stan-

dard GARCH-MIDAS model to characterize the different responses of the

stock market for short- and long-term horizons, separately or in combination.

The unique timespan of nearly 100 years of the Dow Jones Industrial Average

(DJIA) daily returns allows us to understand the stock market volatility under

extreme shocks from a historical perspective. The in-sample empirical results

clearly show that the DJIA stock volatility is best fitted to the GARCH-

MIDAS-SLES model by including the short- and long-term impacts of extreme

shocks for all forecasting horizons. The out-of-sample results and robustness

tests emphasize the significance of decomposing the effect of extreme shocks

into short- and long-term effects to improve the accuracy of the DJIA volatility

forecasts.

KEYWORDS

extreme shocks, GARCH-MIDAS, stock volatility, volatility forecasting

1|INTRODUCTION

This article investigates the importance of extreme

shocks (e.g., war, political crisis, and the global financial

crisis) to stock volatility forecasting and proposes

new extended generalized autoregressive conditional

heteroskedasticity–mixed data sampling (GARCH-

MIDAS) models for stock volatility to characterize the

impact of extreme shocks divided by short- and long-term

components. With this new modeling framework, we can

provide valuable information about the predictive power

of extreme shocks on stock volatility forecasting accuracy.

More importantly, the new models contain the impacts of

extreme shocks on different time ranges, which can

enhance the accuracy of stock price volatility forecasts.

The notion of the significant effect of extreme shocks

on stock volatility is not new. Based on previous studies

(see, e.g., Chou, 2005; Liljeblom & Stenius, 1997;

Schwert, 2011), the existence of the effect of extreme

shocks on stock volatility is universally recognized. From

an economic point of view, short- and long-term traders

both exist in the stock market. Short-term traders are

looking for a quick way to take advantage of short-term

fluctuations in the market (Otchere & Chan, 2003), while

long-term traders are waiting for the stock's value to rise

over an extended period of time (Greenwood, 2005),

which requires patience and the ability to overlook short-

term fluctuations in value. According to Lee, Jiang, and

Indro (2002), who have shown how market sentiment

has a systematic impact on stock volatility, stock market

Received: 10 August 2019 Accepted: 12 January 2020

DOI: 10.1002/for.2668

Journal of Forecasting. 2020;39:797–810. wileyonlinelibrary.com/journal/for © 2020 John Wiley & Sons, Ltd. 797