Can technical indicators predict the Chinese equity risk premium?

• Published date: 01 March 2022
• Author: Mingwei Sun,Paskalis Glabadanidis
• Date: 01 March 2022
• DOI: http://doi.org/10.1111/irfi.12344

ORIGINAL ARTICLE

Can technical indicators predict the Chinese

equity risk premium?

Mingwei Sun

1

| Paskalis Glabadanidis

2

1

Australian Institute of Business, the

University of Adelaide, Adelaide, South

Australia, Australia

2

Adelaide Business School, the University of

Adelaide, Adelaide, South Australia, Australia

Correspondence

Paskalis Glabadanidis, Accounting and

Finance, Adelaide Business School, the

University of Adelaide, Adelaide, SA 5000,

Australia.

Email: paskalis.glabadanidis@adelaide.edu.au

Abstract

We find that technical indicators have substantial predic-

tive power over the Chinese equity risk premium. Techni-

cal indicators complement macroeconomic variables in

predicting the Chinese equity risk premium. The predic-

tive power is more pronounced at a weekly frequency

rather than a monthly frequency as suggested by the out-

of-sample tests. Furthermore, weekly-level technical indi-

cators can predict the firm-level excess returns while

monthly-level indicators cannot. The weekly-level indica-

tors can also predict sorted portfolio excess return and

risk factors. Overall, in comparison with the US stock

market, the Chinese stock market seems to have higher-

frequency price trends. The cross-sectional predictive

power of the technical indicators is closely related to

market capitalization rather than volatility.

KEYWORDS

equity risk premium predictability, macroeconomic variables,

momentum, moving averages, out-of-sample forecasts, short-

term trend, technical analysis, the Chinese stock market

JEL CLASSIFICATION

C22; C53; E32; G11; G12; G17

This paper has benefited from discussions with and comments from Syed (Aku) Zemin Ali, Juan (Jane) Luo, Patrick Roger, LiminXu, as well as seminar

participants in the University of Adelaide Business School research seminar. The authors would also like to thank the editor (Dragon Tang) as well as an

anonymous referee for their valuable comments on our earlier draft. This research did not receive any specific grant from funding agencies in the public,

commercial, or not-for-profit sectors.

Received: 4 April 2019 Revised: 2 November 2020 Accepted: 12 January 2021

DOI: 10.1111/irfi.12344

© 2021 International Review of Finance Ltd. 2021

114 International Review of Finance. 2022;22:114–142.wileyonlinelibrary.com/journal/irfi

1|INTRODUCTION

Numerous studies investigate the ability of macroeconomic variables to predict the U.S. equity risk premium

(e.g., Breen, Glosten, & Jagannathan, 1989; Campbell, 1987; Cochrane, 2007; Fama & French, 1988; Fama &

French, 1989; Fama & Schwert, 1977; Ferson & Harvey, 1991; Lettau & Ludvigson, 2001; Pástor &

Stambaugh, 2009; Pettenuzzo, Timmermann, & Valkanov, 2014), as well as many international stock markets

(e.g., Ang & Bekaert, 2006; Bekaert & Hodrick, 1992; Cutler, Poterba, & Summers, 1991; Ferson & Harvey, 1993;

Harvey, 1991; Henkel, Martin, & Nardari, 2011; Hjalmarsson, 2010; Solnik, 1993), including China (e.g., Chen, Jiang,

Li, & Xu, 2016; Chen, Kim, Yao, & Yu, 2010; Goh, Jiang, Tu, & Wang, 2013; Jiang, Rapach, Strauss, Tu, &

Zhou, 2011).

Unlike macroeconomic variables, technical indicators have received far less attention despite their documented

profitability and predictive power (e.g., Glabadanidis, 2015; Han, Yang, & Zhou, 2013; Han, Zhou, & Zhu, 2016;

Neely, Rapach, Tu, & Zhou, 2014; Park & Irwin, 2007; Smith, Wang, Wang, & Zychowicz, 2016), as well as their pop-

ularity with practitioners (Covel, 2009; Lo & Hasanhodzic, 2010; Lo & Hasanhodzic, 2011; Menkhoff, 2010; Men-

khoff & Taylor, 2007; Park & Irwin, 2007; Schwager, 2012). Technical indicators are based on past price and volume

patterns and have been used to identify price trends that may persist into the future. We examine whether technical

indicators can predict the Chinese equity risk premium. First, we add to the extant literature on equity risk premium

predictability by investigating whether technical indicators in conjunction with macroeconomic variables can predict

the Chinese equity risk premium (for a recent example, see Neely et al., 2014). Second, while the technical analysis

literature is focused mostly on the profitability of technical indicators (e.g., Brock, Lakonishok, & LeBaron, 1992; Lo,

Mamaysky, & Wang, 2000), we focus on the ability of technical indicators to predict the equity risk premium.

Theoretically, macroeconomic variables can predict the equity risk premium because they measure changing

macroeconomic conditions which are the fundamental drivers of time-varying investment opportunity sets. This pre-

dictive ability is consistent with rational asset pricing and reflects fluctuations in aggregate risk exposure which pro-

duces time-varying discount rates (see, e.g., Cochrane, 2011; Rapach & Zhou, 2013).

In contrast, the underlying reason for the predictive ability of trend-following technical indicators is still unclear

and controversial. In terms of a rational asset pricing model, Cespa and Vives (2012) show that, in the presence of

heterogeneous information, asset prices can systematically diverge from fundamental values and generate rational

price trends. Alternatively, behavioral biases can also lead to price deviations from fundamental values. For example,

Hong and Stein (1999) show that investors tend to underreact to new information at first and then overreact in the

long run. Daniel, Hirshleifer, and Subrahmanyam (1998) suggest that investors are overconfident about their private

information and overreact to confirming news, therefore causing a trend in market prices. Barberis, Shleifer, and

Vishny (1998) posit that if investors give less weight to new information then this causes a market price continua-

tion. Another important factor that appears to be related to the predictive power of technical indicators is investor

sentiment. Smith et al. (2016) show that, during high-sentiment periods, hedge funds using technical analysis perform

better, have lower risk, and exhibit superior market-timing ability when compared to hedge funds that do not use

technical analysis. Neely et al. (2014) find that technical indicators can significantly predict changes in investor senti-

ment, while a variety of investor sentiment proxies are correlated with the U.S. and international stock returns

(Baker & Wurgler, 2006; Baker & Wurgler, 2007; Baker, Wurgler, & Yuan, 2012).

Three features of the Chinese stock market may contribute to the predictive power of technical indicators in

China. First, the Chinese stock market is dominated by individual investors who are prone to behavioral biases. By

March 2018, 99.73% of the total security accounts belong to individual investors (China Securities Depository and

Clearing, http://www.chinaclear.cn), who contribute to 82.01% of the total trading volume and hold 21.42% of the

total market capitalization in 2017 (Shanghai Stock Exchange Statistics Annual, http://www.sse.com.cn/aboutus/

publication/yearly/). For this reason, the Chinese retail investors play a crucial role in providing liquidity and in deter-

mining the stock returns. More importantly, many of these retail investors are not financially sophisticated, gambling

oriented, and prone to behavioral biases. Therefore, as discussed above, the potential prevalence of behavioral biases

SUN AND GLABADANIDIS 115