See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/342644550

Overreaction Effect: Evidence from an Emerging Market (Shanghai Stock

Market)

Article  in  International Journal of Managerial Finance · July 2020

DOI: 10.1108/IJMF-01-2019-0033

CITATIONS READS

13 438

4 authors:

Dr Krishna Reddy Muhammad Ali Jibran Qamar

Toi Ohomai Institute of Technology/Te Pukenga COMSATS Secretariat
99 PUBLICATIONS   1,711 CITATIONS    23 PUBLICATIONS   174 CITATIONS   

SEE PROFILE SEE PROFILE

Nawazish Mirza Fangwei Shi

Excelia Business School The University of Waikato
104 PUBLICATIONS   2,369 CITATIONS    1 PUBLICATION   13 CITATIONS   

SEE PROFILE SEE PROFILE

Some of the authors of this publication are also working on these related projects:

Market timing of security offerings: International evidence View project

Call for Papers - Energy Economics - Green Finance View project

All content following this page was uploaded by Dr Krishna Reddy on 27 September 2020.

The user has requested enhancement of the downloaded file.

 The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/1743-9132.htm

Overreaction effect: evidence from Overreaction

effect of
an emerging market (Shanghai Chinese stock
market
stock market)
Krishna Reddy
Postgraduate Business, Faculty of Health, Education and Environment,
Rotorua, New Zealand Received 18 March 2019
Revised 16 July 2019
Muhammad Ali Jibran Qamar 7 July 2020
Accepted 7 July 2020
Department of Management Sciences,
COMSATS University Islamabad - Lahore Campus, Lahore, Pakistan
Nawazish Mirza
La Rochelle Business School, La Rochelle, France, and
Fangwei Shi
School of Accounting, Finance and Economics, Waikato Management School,
University of Waikato, Hamilton, New Zealand

Abstract
Purpose – The purpose of the study is to examine overreaction effect in the Chinese stock market after the
global financial crisis (GFC) of 2007 for all the stocks listed in Shanghai Stock Exchange (SSE) Composite
50 index.
Design/methodology/approach – To capture overreaction effect in the stock listed at SSE 50 Index, a time
series analysis of average cumulative abnormal return within a unified framework is applied for the period of
January 2009 to December 2015. From these loser and winner portfolios, contrarian strategy is applied to build
arbitrage portfolio, which is the difference of mean reversions between loser and winner portfolios. The
portfolio construction is based on a 12-month formation period and 6-month testing period for intermediate-
term analysis and. for short-term analysis, 6 month formation and 3 month testing periods. The authors also
applied regression analysis to test a return reversal effect for the sampled period.
Findings – Results show that contrarian strategy yields positive excess returns for the arbitrage portfolio for
most of the testing periods. The intermediate baseline case shows the arbitrage portfolio producing an average
excess return of 14.1%, while even the short-term one produces 4%, which is statistically significant at the 5%
level. The study finds asymmetrical overreactions in the SSE especially for loser portfolios. The biggest winner
and loser portfolios follow the mean reversal effect. Moreover, before-after test for the biggest winner and loser
portfolios shows that the losers recovered and beat the market immediately.
Practical implications – The study could benefit government, policy makers and regulators by studying how
presence of more individual investors than institutional investors of China stock market leads to more irrational
decisions giving rise to volatility. The regulators could build favourable policies for institutional investors to give
them incentive to invest more than individual investors through which market volatility could be controlled.
Originality/value – This research contributes to market behaviour research, showing how working under
hypotheses of overreaction; gains can be made with contrarian investment strategy through arbitrage
portfolios. The authors provide specific additional support for the short and medium-term overreaction in the
SSE for the period 2009–2015 using regression analysis.
Contribution to Impact – This research contributes to market behaviour research, showing how working
under hypotheses of overreaction; gains can be made with contrarian investment strategy through arbitrage
portfolios. We provide specific additional support for the short and medium-term overreaction in the SSE for
the period 2009–2015 using regression analysis.
Keywords Overreaction, Momentum, Shanghai Stock Exchange (SSE)
Paper type Research paper

International Journal of Managerial

Finance
© Emerald Publishing Limited
1743-9132
JEL Classification — G14 DOI 10.1108/IJMF-01-2019-0033
IJMF 1. Introduction
The existence of anomalies like a mean reversal in the stock markets has been well
documented in the developed market literature. It highlights the volatility and
unpredictability of the equity market. Return reversal effect is one anomaly that brings
researchers to question the reliability of the efficient-market hypothesis (EMH) (Rosenberg
et al., 1985; Chan, 1988; Lehmann, 1990; Lo and MacKinlay, 1990). According to EMH, the
stock prices reflect all possible market information, and therefore, earning a significant
abnormal return is impossible (Basu, 1977). With return reversal effect, an investor could earn
a significant abnormal return, which is close to the weak form of an EMH. According to return
reversal effect, the preceding period’s poorly performed stock (losers) performs better than
the previous period’s best-performed stocks (winners) in the following period which
contradicts with the EMH (Zhang et al., 2018; Malkiel, 2003; Zarowin, 1990). This abnormal
return earning behaviour is referred to as the “overreaction” phenomenon.
The overreaction phenomenon suggests that due to the new information; the investors
overreact in the initial period, leading the prices to deviate from its fundamental values and
later on correct by bringing the prices back to the fundamental values. Initially, De Bondt and
Thaler (1985) showed that buying stocks that have been losers (worst-performance) and
selling that have been winners (best-performers) over the previous 3-to-5 years could generate
excess profitability over the next 3-to-5 years. They called this method of generating
abnormal return “Contrarian Strategies”. Later, Fama and French (1988), Poterba and
Summers (1988) and Cutler et al. (1991) showed that stock prices reverse their trend lines over
long horizon period, that is, winners in the past may become losers in the future and vice
versa. Chopra et al. (1992), Richards (1997) and Balvers et al. (2000) tested the mean reversion
and investigated the profitability of contrarian strategies. Maheshwari and Dhankar (2015)
speculated the long course reversal effect in the Indian stock market and reported a strong
inference of reversal effect in India with cumulative abnormal return strategy.
A plausible interpretation of mean reversion is that the stock markets consistently
overreact to new information and drive the winner and loser stocks to the opposite positions.
In the USA, individual stock returns exhibit a short-term reversal effect within a few days or
weeks, a medium-term momentum effect in 3 months to 1 year and a long-term reversal effect
within 3–5 years (Zhang et al., 2018). The short-term mean reversal effect focuses on reactions
to unexpected information. The long-term mean reversal effect studies indicate stock prices
indeed temporarily swing away from their fundamental value due to both optimistic and
pessimistic news (Li et al., 2014). According to De Long et al. (1990), the existence of
“optimistic” traders can push stock prices away from fundamental values. Rational trader
may jump on the bandwagon and purchase (sell) ahead of the “optimistic” traders upon
hearing good (bad) news, which leads to short-term momentum. Thus, market overreacts to
news asymmetrically, the loser overreacting more than the winner.
The concept of emerging markets is described as the countries with specific
characteristics with less developed equity markets, less liquid, less industrialised
(Schoenfeld, 2011), higher transaction costs but also characterised with high growth
potential and more intended for economic liberalisation (Bekaert et al., 2007). Due to
differential behaviour of stakeholders in an emerging market, they need a different criterion
to deal with market information. Fama and French (1996) document that emerging markets
have a value premium in their stock returns. This led researchers to examine emerging
markets with specific characteristics. The Chinese stock markets were established in the
1990s and had been growing rapidly in terms of the trading volume, companies and market
capitalisation. These are the fourth largest in the world, with over US$5 trillion market
capitalisation. Chinese stock market is the fastest growing in the emerging world, i.e. between
November 2014 and June 2015 Shanghai Stock Exchange (SSE) composite index rose by
150% and fell by 40% in the following three months. The Chinese stocks markets could be
characterised by high returns and excessive volatility, as compared to stocks traded in Overreaction
developed markets such as the US and the UK. Thus, fast growth and excessive volatility effect of
warrant an investigation from investors, practitioners and policymakers’ viewpoint. This
ranks China ahead of most EU members other than the UK, concerning the significance of the
Chinese stock
stock market to the economy. Chinese stock markets are attracting foreign investor attraction market
due to the prevalence of enormous growth opportunities (Wang and Chin, 2004). China has
the largest population (1.2bn) in the world, having nominal GDP ($1.02 trillion), which is the
second largest in Asia after Japan. The GDP growth rate of 8% per annum is highest in
the world.
China is in the race of economic growth and the process of development. Unlike the
developed economies, the returns reversal effect in the emerging market of China is
controversial as prior empirical studies from Shanghai and Shenzhen stock markets have
reported mixed results. For example, Malin and Bornholt (2013) and Noar (2011) analysed the
A-share stocks listed on the SSE (SSE) and reported that there is a long-term winner-loser
return reversal effect in China. Earlier, Kang et al. (2002) investigated all Chinese A-stocks and
reported that there is a short-term return reversal in Chinese A-stock market, whereas other
empirical studies claimed that future market returns are not dependent on the past
performance. However, empirical research is not limited to finding the return reversal effect in
stocks, i.e. Zhao et al. (2016) explored the determinants of the return reversal and found
liquidity to be the most important factor that causes the return reversal effect. Naughton et al.
(2008) showed that there is a momentum effect in the Chinese stock market, whereas Li et al.
(2010) observed no such momentum effect in the Chinese market. These mixed findings,
reported for return reversal effect in China, motivate this study. Shumway and Wu (2006)
document disposition effect as a driving factor of momentum effect in the Chinese stock
market. Reddy et al. (2019) examined return reversal effect in the Shanghai A-share market.
With the application of late-stage contrarian strategy of Malin and Bornholt (2013) for the
period March 2011-2016, they demonstrate the presence of long-term reversal effect in the
Shanghai A-share market. They also demonstrate short term momentum effect. Further, they
document portfolios having small size and lower book to market values explains a large
portion of return reversal effect.
Rare work is found in China who covers overreaction effect in the account of post-financial
crises horizon. In the fastest growing and emerging market of China, how the news
information of global financial crisis (GFC) of 2007 affects Chinese investors creating an
overreaction effect in workable to study. Most of the studies worked on collectively Asian
markets or emerging market countries, the study on China alone for seeking mean reversion
and momentum existence financial crises over-reaction effect is still an open question. Zijun
and Lin (2015) studied the overreaction in the Shanghai stock market from 2004 to 2014. Their
results explored that after 2007, the effects of overreaction start decreasing. They indicate the
changing behaviour of overreaction to be due to the global financial crisis. Sharma et al. (2015)
documented the presence of herding behaviour of investors in pre- and post-financial crises in
both Shanghai and Shenzhen stock market. They showed the evidence of asymmetric
herding behaviour with greater magnitude of herding behaviour on up markets than on down
markets. Lim et al. (2008) examined pre and post financial crises overreaction effect of in
emerging markets of Asia. They showed that most of the Asian markets in post-financial
crises improved their market efficiency. They indicate that investor not only reacts to local
news but also the news originating in other markets leading to more significant overreaction
effect in post-financial crises period.
The current research contributes to the academic literature in several ways. First, as
mentioned before, although the occurrence of reversal effect is studied in the emerging
market of China, they all reported mixed results. China remains the most important emerging
market for exploration because China is one of the fastest growing economies in the world
IJMF and remains one of the few countries that have a negative correlation with US stock market
(Kang et al., 2002). Since the revolution of stock markets in 1990, the total capitalisation of
Chinese markets has increased to more than 50% of country GDP in 2000 (MacNeil, 2002).
Further, it has more asymmetries due to limited information and high transaction costs
(Harvey, 1995).
Contrary to the developed markets where large institutional investors invest in stock
markets, around 85% investors in Chinese stock market are inexperienced individuals who
trade more frequently than foreign counterparts, i.e. average stock holding period in China is
two months (Zhang et al., 2018). According to Nofsinger and Sias (1999), individual investors
made more irrational decisions than institutional investors. These investors’ irrational
behaviour, market volatility, high transactions costs due to the short holding period and
asymmetric information offer a strong case for investigating overreaction in the SSE 50
index. Additionally, previous researchers analysed the generation of abnormal returns
through contrarian and momentum-based trading strategies covering developed markets
and did not account for after effects of GFC with relevant to tremendous increase in trading
volume and market capitalisation in Chinese stock market.
Second, Chinese stock markets have distinctive features in comparison to the developed
stock markets in terms of government control, institutional structure, liquidity and cultural
background. The accounting system and stock types are different in the SSE where A-shares
are available to mainland investors and B shares available for mainland and foreign
investors. The different characteristics of both types of stocks (A and B) could affect the
intensity and speed of overreaction and mean reversal effect (Wang et al., 2004). Such
differences affect the pattern in stock returns compared with those observed in developed
stock markets. Furthermore, the presence of strict government regulation, transaction cost
and limited investment tools may also affect the significance overreaction effect in this
respect. The study of the China market with distinctive features provides a potential
contribution for emerging markets investors
Third, the study augments the literature by providing sample period that covers January
2009-December 2014 to test the effects of overreaction effect freshly after happening of GFC
on Chinese investors. Further, to analyse the significance of risk-based factors based on the
Fama-French model, the sample is expanded with the use of daily performance instead of
monthly performance to form loser and winner portfolios. Due to China’s enormous growth,
the study of sample test of the long run with ultra-recent short-term return, reversal effect is
required, right after the financial crises which justifies the sample period of January
2009-December 2014.
To capture overreaction effect in the stocks listed in SSE 50 Index, average cumulative
abnormal return methodology of Maheshwari and Dhankar (2015), Tripathi and Agrawal
(2009) and Balvers and Wu (2006) within a unified framework is applied from January 2009 to
December 2014. The study of stock market abnormal returns is incomplete without event
study. The event studies capture the event news effects on stock return. The event studies are
based on time series analysis. As the study adheres to find GFC post news effects on Chinese
investors, the event study (time series analysis) is more appropriate to apply rather than just
performing contrarian methodologies (Lee et al., 2000). The portfolio construction is based on
a 12-month formation period and 6-month testing period for intermediate-term analysis. For
short-term analysis, 6-month formation and 3-month testing periods are taken. As the period
taken for analysis covers the post-GFC in China market, the empirical evidence about
overreaction effect after GFC in China is scarce, and it will be interesting to seek whether the
findings before the crisis still holds after the crisis as well. Results indicate that overreaction
in the loser portfolio is significant for the testing periods of 6 and 3 months. Furthermore, the
biggest winner and loser portfolios indicate mean reversal effect. Our results on mean
reversion support the hypotheses of overreaction and positive-feedback trading framework.
 The remainder of the paper formulated is as follows. Section 2 provides a brief review of Overreaction
the relevant literature. Section 3 pronounces the data and methods used. Section 4 provides effect of
the results and discuss and section 5 presents the conclusion.
Chinese stock
market
2. Literature review
To understand overreaction in the market, it is best first to review the normal reaction. The
most thought-provoking, well-studied and controversial theory of market normality is the
EMH. EMH states that “Markets are rational and efficient,” and “Prices fully reflect all
available information”. According to the EMH, it is impossible to gain abnormal returns
because stocks will always trade at their fair value and will yield average returns based only
on the information available about the past. The EMH has been widely accepted by the
financial economists even up until now and even with much questioning of it.
It provides investors with a rule and equation to calculate returns in the real market, which
is close to the weak form of EMH.
Early studies (Bailey, 1994; Ma, 1996; Su and Fleisher, 1998) document a significant
underpricing in the Chinese equities due to overreaction effect. The overreaction hypothesis
focuses on the market participants rather than the market itself, predicting that the
participants will overreact to new and unexpected events by pulling the stock prices far from
their real value, thus virtually creating the market. At first, it was considered for short
periods, because investors realise their mistakes rapidly and take corrective actions later.
Corrective actions include investing in the opposite direction, and prices revert to their real
fundamental levels. Hence, according to this hypothesis, it is possible to earn abnormal
returns using “contrarian strategy” that is the investors can purchase stocks that have
performed poorly and sell those which have performed well in the past.
Stein (1989) worked on “term structure” of options’ implied volatilities, using data on S&P
100 index options. The study documents volatility to be strong contributor for mean
reverting. They support the rational behaviour of investor and also suggested that long-
maturity options investors tend to “overreact” to changes in the implied volatility of short-
maturity options. Using the late-stage contrarian strategy of Malin and Bornholt (2013) and
Reddy et al. (2019) in their study, find a long-term return reversal effect and a significant
short-term momentum effect in the Shanghai A-stock market stocks. Further, they have
documented portfolios having small size and lower book to market values explain a large
portion of return reversal effect. Maheshwari and Dhankar (2015) speculated the long-term
overreaction effect in the stock market of India and reported a strong inference of reversal
effect in India with cumulative abnormal return strategy. Antoniou et al. (2005) investigate
the existence and sources of contrarian stock market for Athens Stock Exchange market.
They document serial correlation in equity returns, which leads to significant short-run
contrarian profits. Further, they evidenced the reason for contrarian profits are due to the
overreaction to the firm-specific components.
The predictability of earning an abnormal return in the stock market with a contrarian
strategy had received increased attention. Overreaction hypothesis introduced the
consideration of the behaviour of investors into market perspectives, resulting in the
perspective of behavioural finance. Behavioural finance offers alternative explanations for
the deviation of prices from their fundamental values. According to Hirshleifer (2001),
behavioural finance is based on the factors affecting human behaviour relevant to
investment decisions and focuses on factoring the psychological biases to improve
financial decisions. Daniel, Hirshleifer and Subrahmanyam (1998) assume that investors
are overconfident about their information or experience and that, due to self-attribution
bias, this overconfidence keeps on increasing and leads to an overreaction in their
investments. Mean reversions are not the only anomalies leading to overreaction effect.
IJMF The other various market anomalies giving rise to overreaction effect are well examined
by previous researchers including risk mismeasurement (Chan, 1988), bid-ask spread
(Atkins and Dyl, 1990), size effect (Zarowin, 1990) and biases in computed return (Conrad
and Kaul, 1993; Dissanaike, 1994).
De Bondt and Thaler (1985) introduced “winner” and “loser” portfolios and argued that
people systematically overreact to unexpected news. They suggested subsequent price
movements will follow two hypotheses, i.e. (1) extreme movements in stock prices in the
opposite direction and (2) the more extreme the initial price movement, the higher will be the
subsequent adjustment. From here, the extreme movements are described as investor
reaction based on new market information leading to the high fluctuation in stock prices due
to information asymmetry. The higher the fluctuation in stock prices, the greater would be
the deviation from fundamental values and the higher the adjustment time where prices
revert to their original values. The overreaction effect has been confirmed in developed
countries like the UK (Clare and Thomas, 1995), Spain (Alonso and Rubio, 1990), New Zealand
(Bowman and Iverson, 1998) and Germany (Lobe and Rieks, 2011), with a variety of formation
and testing periods. Researchers also focused on short-run overreaction (daily and weekly
data from a one-week testing period) (Ahmed and Hussain, 2001; Bowman and Iverson, 1998)
whereas others use more extended testing periods of one or two years (Daniel and
Hirshleifer, 1998).
For the Spanish stock market, intense overreaction was observed from 1967–1984 to
1963–1997, which gets stronger when longer formation and testing periods are used (Alonso
and Rubio, 1990; Forner and Marhuenda, 2004). Stock (1990) for Germany, Swallow and Fox
(1998) for New Zealand and Dubois and Bacmann (1998) for France confirmed the presence of
overreaction effect and reported that contrarian strategy leads to smaller yet significant
profits. However, Brailsford (1992) for Australia and Kryzanowski and Zhang (1992) for
Canada found weak evidence of the overreaction effect with one, two, three, five and eight
years testing and formation periods. Similarly, Clare and Thomas (1995) found weak
overreaction effect in the stock market of the UK from 1955 to 1990, and these abnormal
returns were due to the size effect.
Richards (1997) using total returns of sixteen market indices of loser and winner portfolios
examined common international risk factors that were well integrated into the markets. He
documented that for three to four years, losers outperformed winners. Similarly, Baytas and
Cakici (1999) examined seven stock markets (the US, Canadian, Japanese, French, Italian,
German and the UK) and found strong overreaction effect in two- and three-years periods for
all countries except the US and Canada stock markets. These results are indicating the
significant implication of overreaction effect in developed economies but with varying
intensity.
Studies also explored the overreaction effect on developing countries. Da Costa (1994)
found an overreaction effect in the Brazilian stock market, but it was asymmetrical, which
means that only the winner portfolios reversed. Similarly, Locke and Gupta (2009) and
Tripathi and Gupta (2009) reported the overreaction effect in Sri Lankan and Indian stock
markets, respectively. Ahmad and Hussain (2001) and Ali et al. (2011) reported on the
overreaction effect and seasonality in Malaysia. They highlighted that overreaction was
significant before the Asian financial crises of 1997 and had gradually diminished and
become insignificant since then.
In China, most of the studies related to overreaction effect are examined without
consideration of the financial crises of 2007. Amongst these studies, Li et al. (2010) examine 25
momentum and contrarian trading strategies using monthly stock returns in China for the
period from 1994 to 2007. Their results indicate that no momentum profitability in any of the
25 strategies exists. However, there was evidence of reversal effects where the past winners
become losers and past losers become winners afterwards. The contrarian profit strategy
significantly holds for the short-term formation and holding periods 1–3 months. They Overreaction
indicate that contrarian strategies could generate about 12% return per annum. Kang, Liu effect of
and Ni (2002) examined short-term contrarian reversal effect and intermediate term
momentum effect in China using weekly data from 1993 to 2000. They explained the leading
Chinese stock
cause of abnormal return to be investor overreaction to firm-specific information, but this did market
not hold for intermediate-term momentum effect.
Wu (2011) explored the contrarian and momentum strategies in SSE from 1990 to 2001.
They found that contrarian strategies with significant overreaction effect provide more
abnormal returns as compared to momentum strategies. Wang and Chin (2004) examined
mean reversal and momentum effect in China stock market. They report past trading volume
to be the main driving factor of mean reversal and momentum effect. They also document
market characteristics such as the dominance of individual investors and prohibition on short
sales may explain these abnormal stock returns. Chan et al. (2000) argue that firm-specific
events are drivers of the momentum effect in China. The investors under-react to the specific
event, or there may be delayed overreaction to information due to positive feedback trading.
To cover mean reversal effect in post-financial crises horizon very few studies are here to
mention where most of the studies worked on collectively Asian markets or emerging market
countries, the study on China alone for seeking mean reversion and momentum existence
after financial crises is still an open question. Zijun and Lin (2015) studied the overreaction in
the Shanghai stock market from 2004 to 2014. Their results explored that after 2007, the
effects of overreaction start decreasing. They indicate the changing behaviour of
overreaction to be due to the global financial crisis. Susan et al. (2015) documented
presence of herding behaviour of investors in pre and post financial crises in both Shanghai
and Shenzhen stock market. They showed the evidence of asymmetric herding behaviour
with greater magnitude of herding behaviour on up markets than on down markets. Lim et al.
(2008) examined pre and post financial crises overreaction effect of in emerging markets of
Asia. They showed that most of the Asian markets in post-financial crises improved their
market efficiency. They indicate that investor not only reacts to local news but also the news
originating in other markets leading to more significant overreaction effect in post-financial
crises period.
Banz (1981) introduced size as an essential factor in affecting returns of security,
indicating that stocks with a lower market capitalisation (small stocks) tend to have higher
average returns. Fama and French (1996, p. 2012) revealed that the book explains maximum
variation in stock returns to market equity and size. The returns on small stocks are more
sensitive to the risk captured by the size factor than the returns on big stocks that emphasise
that the profitability of the long run return reversal effect is associated with size risk. Zarowin
(1990) depicted that past loser portfolios are dominated by small size stocks with higher risks
that generate a higher return in longer time horizons compared to past winner portfolios of
big size stocks and vice versa. Past winner, portfolios with big size stocks of large-scale
companies could become loser portfolios.
With this size effect explained above, since 50 largest companies of China are listed on the
Shanghai A-Stock Exchange 50 index compared to the Shenzhen listing, we assume return
reversal effect would be present in the Shanghai market. Therefore, this research explored the
presence of a weak form of EMH with a mean reversal effect from 2009 to 2014 in SSE with
consideration of size effect as well as post financial crises after effects. The sample includes 50
constituent stocks in the SSE 50 Index. These 50 largest firms represent the highest market
capitalisation. This research intends to explore whether mean reversal effect prevails for
large firms after the global financial crisis in China. If results explored the significant
implication of overreaction, then it can be argued that it is not due to only size effect but also
due to financial crises or specific market characteristics. Thus, the contribution of this
IJMF research will be its investigation of the overreaction for large size firms after GFC in case of
Chinese listed firms. The literature review led us to the development of a hypothesis.
H1. Overreaction effect exists in the Shanghai A stock market in post-financial crises
period.

3. Data and methodology

The data for this study consists of the monthly returns of all 50 constituent stocks in the SSE
50 Index in the SSE from January 2009 to December 2014 (part of the sample extends back to
2006 for relevant purposes). The market return is taken as the total return of the SSE shown in
the index. Following the methodology of (Maheshwari and Dhankar, 2015; Tripathi and
Agarwal, 2009; Balvers and Wu, 2006), the time series analysis of cumulative abnormal
return is applied for construction of loser and winner portfolios. From these loser and winner
portfolios, contrarian strategy is applied to build arbitrage portfolio, which is the difference of
mean reversions between loser and winner portfolios. In this contrarian strategy, formation
and holding periods are formed for winner and loser portfolios construction. The portfolios
construction is based on 12 months formation period and 6 months’ testing period for
intermediate-term analysis. For short-term analysis, 6 months formation period and
3 months’ testing periods are taken. The study employed overlapping portfolios where
portfolios are rebalanced at the start of each year.

3.1 Excess cumulative return

At first, the stock price data are converted into simple percentage returns as follows where Ri;t
is the monthly return, Pi;t is the price on month t and Pi;t−1 is the price on month t-1.
Pi;t  Pi;t−1
Ri;t ¼
Pi;t−1

Beginning in January 2009 to December 2014, the cumulative market return of stock i in a
period of t months (CRi;t ) is calculated as follows:
Y t
CRi;t ¼ ð1 þ Ri;j Þ ¼ ð1 þ Ri;1 Þð1 þ Ri;2 Þ: : :ð1 þ Ri;t Þ (1)
i¼1

Where Ri;j is the monthly return of stock i in month j. Monthly stock returns from the period of
January 2009 to December 2015 in the index is calculated for both the 12-months formation
and 6-months test periods in intermediate-term analysis and 6-months formation and
3 months test periods for short-term analysis. The cumulative returns for t months in the
testing periods are the buy-and-hold returns for the period of t months. We employed a
geometric mean instead of arithmetic mean because the former can reduce the error caused by
bid-ask spread, as stated by Conrad and Kaul (1993). Similarly, Dissanaike (1993) argued that
calculation with arithmetic mean is an inaccurate method in computing multi-period returns
from single period returns as the strategy involves rebalancing to equal weights in every
single period. The excess cumulative return ECRi;t for stock i in a period of t month is defined as;
ECRi;t ¼ CRi;t  CRm;t (2)

Where, CRm;t is the total return of the market in the corresponding period of t month. To
calculate excess cumulative return ECR, six years returns have been used. The formation
period is 12 months and testing period is 6 months for intermediate term analysis whereas
6-months formation period and 3-months test periods for short term analysis. The formation
periods are sorted based on their excess cumulative return ECRi;t. Based on excess cumulative Overreaction
returns, the monthly cumulative returns of 12-months formation period and separate 6- effect of
months formation period are sorted in descending order. The winner portfolios would be
regarded to those stocks having the highest excess cumulative returns whereas the loser
Chinese stock
portfolios are those having the lowest value of excess cumulative returns. The winner portfolio market
contains the top 5th percentile stocks with the highest ECR returns, termed as ECRW ;T . The
loser portfolio contains the bottom 5th percentile stocks with the lowest ECR termed
as ECRL;T .

3.2 Average excess cumulative return

For intermediate-term analysis and short-term analysis in each testing period T, the average
of excess cumulative returns of top 5th percentile stocks are taken termed as AECRW ;T and an
average of excess cumulative return of bottom 5th percentile stocks AECRL;T are
constructed. In intermediate term analysis the AECR for all portfolio securities are
calculated for each of the next 12 months formation period (F) and 6 months testing periods
(t). In short term analysis, the formation periods (F) are 6 months and testing periods (t) are
3 months. The arbitrage portfolio as described above is the difference between loser and
winner portfolios that is AECRL;T − AECRW ;T ¼ DL;T−W ;T is calculated using following
equations respectively:
3.2.1 Intermediate-term analysis.
PN
j¼1 ECRw;i;t
AECRW ;T ¼ ; F ¼ 1; 2; 3 . . . 12; t ¼ 1; 2; 3 . . . 6 months (3)
N
PN
j¼1 ECRL;i;t
AECRL;T ¼ F ¼ 1; 2; 3 . . . 12; t ¼ 1; 2; 3 . . . 6 months (4)
N
DL;T−W ;T ¼ AECRL;T  AECRW ;T (5)

3.2.2 Short-term analysis.

PN
j¼1 ECRw;i;t
AECRW ;T ¼ F ¼ 1; 2; 3 . . . 6; t ¼ 1; 2; 3 months (6)
N
PN
j¼1 ECRL;i;t
AECRL;T ¼ F ¼ 1; 2; 3 . . . 6; t ¼ 1; 2; 3 months (7)
N
DL;TW ;T ¼ AECRL;T  AECRW ;T (8)

Where A ECRW ;T and AECRL;T are the average excess cumulative return. ECRW ;i;T is the
excess return of stock i in the winner portfolio, ECRLi;T is the excess return of stock i in the
loser portfolio. DL−W ;T is the return of the arbitrage portfolio.
3.3 Grand average excess return
The excess returns of the winner and loser portfolios of all testing periods are combined and
averaged to obtain the grand average excess returns of the winner portfolios GAECRW and of
the loser portfolios GAECRL as follows. This is done separately for intermediate term
analysis and for short term analysis.
1X n
GAECRW ¼ AECRW ;T (9)
n T¼1
IJMF 1X n
GAECRL ¼ AECRL;T (10)
n T¼1
DGL−W ¼ GAECRL  GAECRW (11)

Where, DGL−W is the grand average return of the arbitrage portfolio. The t-test is applied to
find whether, GAECRW , GAECRL or DGL−W is significant in all testing periods being non-
overlapping. The cumulative returns for t months in the testing periods are the buy-and-hold
returns for the period of t months. Only the beginning and ending prices are needed to
calculate these cumulative returns. The method is consistent with Equation (1).
3.4 Hypothesis
If the mean reversal effect exists in SSE 50 Index stock, then during the testing period (t), the
GAECR of losers must be greater than zero while the GAECR of winners must generate
negative returns because the overreaction hypothesis predicts reversals in returns of past
losing and winning stocks. Hence, it could be stated if the GAECR of arbitrage (DG) portfolio
that is (GAECR (L) – GAECR (W)) is greater than zero then it suggests the presence contrarian
profits in intermediate-term and short-term period analysis. The profitability of contrarian
strategies could be explained with average GAECR of the arbitrage portfolio (DG). As
contrarian strategy suggests a long position in past losers and a short position in past winners,
any positive return in the arbitrage portfolio predicts the profitability of contrarian strategy in
the Chinese stock market. Based on this, three hypotheses of mean reversal are tested:
H1. The grand average excess cumulative return is less than equal to 0.
H2. The grand average excess cumulative return is greater than equal to 0.
H3. Loser minus winner grand average excess difference is greater than equal to 0.
H1 : GAECRW ≤ 0
H2 : GAECRL ≥ 0
H3 : DGL−W ≥ 0

The test statistics were calculated as follows:

GAECRW
tW ¼ .pffiffiffiffi
SW N

GAECRL
tL ¼ .pffiffiffiffi
SL N

DGL−W
tD ¼ .pffiffiffiffi
SD N

Where, SW , SL and SD are the standard deviations of winner portfolio, loser portfolio and
arbitrage portfolio return for testing periods.

4. Results
4.1 Intermediate-term analysis
Table 1 reports the average excess returns of the loser, winner and arbitrage portfolios for
intermediate-term analysis with 12 months formation and 6 months testing periods. As the
testing period is 6 months of five overlapping years, it gives 6 3 10 5 60 months. Therefore, a Overreaction
total of ten tests held in testing periods. The arbitrage portfolio return is positive in test 1 effect of
which means in the next 6 months after the formation period the loser portfolio outperform
the winner portfolio, but the number is not significantly higher than zero, and in test 2, it
Chinese stock
drops to negative then recovers again in the next test. Because of the longer formation and market
test period, the volatility of winner and loser portfolios seems to be around zero and slightly
decreasing (increasing) for the winner (loser) portfolios in tests 8 and 10. Figure 1 shows the
graph of Table 1. The winner portfolios showing a downward trend after the formation
period, the highest return of AECRW in test periods is 0.083441(test 5) which is much lower
than formation one (1.16402) and the lowest return of AECRL in test period is 0.0946 which
is higher than the formation one (0.3831) and reaches a peak of 0.68 in test 10. Overall Test
1,3,4,7,8 and 10 confirms the presence of mean reversion that is contrarian profits in China
stock market SSE 50 index, whereas test 2,5,6 and 9 confirms the presence of momentum
effect.

4.2 Short-term analysis

Table 2 shows the average excess returns of loser, winner and arbitrage portfolios in the
3 months testing period (Formation period: 6 months and testing period 3 months), which
gives 22 tests with six overlapping years. The arbitrage portfolio returns (0.025) is positive
in test 1 which means in the next three months after formation period the loser portfolio
outperform the winner portfolio, but the number is not significantly higher than zero, and
in test 2, it drops to negative then recovers in the next test. The winner portfolios return
(0.07204) keep outperforming the market in test 1 but lower than loser portfolio returns
(0.09743). In test 3 to test 16, the winner portfolios appear to have volatility around zero,
and it drops under zero for test 16 and tests 21 and recovers slightly above zero in test 22.
Figure 2 shows the graph of Table 2. The winner portfolios suffer an average decrease
after formation period, the highest return of AECRW in test periods is 0.076 (test 11) which
is much lower than the formation one (0.61897), and the lowest return of AECRL in test
period is 0.0716 which is higher than the formation one (0.3147) and reaches the peak of
0.259104 in test 20. Overall Test 1,3,4,6,8,12,15. . .21 confirms the presence of return

AECRW AECRL DLW

Test period Mean Mean Mean

TEST1 0.039202 0.062999 0.023797

TEST2 0.020995 0.03561 0.0566
TEST3 0.029166 0.056293 0.027127
TEST4 0.00255 0.007472 0.010024
TEST5 0.083441 0.03047 0.11391
TEST6 0.03025 0.09465 0.0644
TEST7 0.017718 0.11838 0.100661
TEST8 0.10543 0.268075 0.373508
TEST9 0.04165 0.08829 0.04664
TEST10 0.47409 0.683826 1.157917
Average 0.04634480 0.09480250 0.14114840 Table 1.
Average excess returns
T statistic 0.922 1.285 1.168 of winner portfolios,
Note(s): AECRW average excess cumulative returns of the winner portfolios after formation; AECRL average loser portfolios and
excess cumulative returns of the loser portfolios after formation; DLW average excess cumulative returns of arbitrage portfolios
arbitrage portfolios (longing loser portfolios and shorting winner portfolios) after formation. *Significant at (testing period of
5% level 6 months)
 1.2
IJMF

0.8

0.4

0.0

-0.4

Figure 1.
Average excess -0.8
cumulative returns of
T1

T2

T3

T4

T5

T6

T7

T8

T9

0
winner portfolios, loser

T1
S

S
TE

TE

TE

TE

TE

TE

TE

TE

TE
portfolios and

TE
arbitrage portfolios
(testing period of AECRW Mean
6 months) AECRL Mean
DLW Mean

AECRW AECRL DLW AECRW AECRL DLW

Test period Mean Mean Mean Test period Mean Mean Mean

TEST_1 0.07204 0.09743 0.02538 TEST_12 0.003582 0.016248 0.012666

TEST_2 0.01553 0.0170 0.03257 TEST_13 0.008809 0.00339 0.0122
TEST_3 0.00970 0.04260 0.03289 TEST_14 0.00311 0.07168 0.06857
TEST_4 0.0216 0.01416 0.03579 TEST_15 0.02709 0.018837 0.045922
TEST_5 0.03742 0.00714 0.03028 TEST_16 0.009431 0.073163 0.063732
TEST_6 0.0382 0.00883 0.04703 TEST_17 0.03678 0.013985 0.050764
TEST_7 0.04023 0.02130 0.01893 TEST_18 0.09873 0.157361 0.256086
TEST_8 0.0368 0.04766 0.08454 TEST_19 0.09662 0.00274 0.093871
TEST_9 0.06355 0.00357 0.05997 TEST_20 0.12459 0.259104 0.383698
TEST_10 0.0088 0.0351 0.02626 TEST_21 0.13207 0.026054 0.158122
TEST_11 0.07672 0.0031 0.07984 TEST_22 0.083372 0.01241 0.09578
Table 2. Average 0.009279 0.0300907 0.0393687
Average excess returns
T statistic 0.700 2.045* 1.657
of winner portfolios,
loser portfolios and Note(s): AECRW average excess cumulative returns of the winner portfolios after formation; AECRL average
arbitrage portfolios excess cumulative returns of the loser portfolios after formation; DLW average excess cumulative returns of
(testing period of arbitrage portfolios (longing loser portfolios and shorting winner portfolios) after formation. *Significant at
3 months) 5% level

reversal effect in 3-months testing periods. Test 2,5,7,9,10,11,13,14 and 22 confirms

presence of momentum effect.

4.3 Intermediate-term analysis

Table 3 shows the results of grand average excess returns for intermediate-term analysis
(Formation period: 12 months, testing period: 6 months) giving ten tests as depicted in
Table 1. Results show that the winner portfolios underperform the market in all the test
periods and loser portfolios only have two negative test periods. The arbitrage portfolio
shows only in the fourth test period winner portfolio beats loser ones. Based on the results
 0.4
Overreaction
effect of
0.3
Chinese stock
0.2
market

0.1

0.0

-0.1
Figure 2.
-0.2 Average excess
cumulative returns of
S 1
S 2
S 3
S 4
S 5
S 6
S 7
8
S _9
S 10
S 11
S 12
S 13
S 14
S 15
S 16
S 17
S 18
S 19
S 20
S 21
22
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_

winner portfolios, loser

TE ST
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
T_
S
TE

portfolios and
arbitrage portfolios
AECRW Mean (testing period of
AECRL Mean 3 months)
DLW Mean

reported in table that the reversal effect occurs after formation period. Figure 3 shows the
graph of Table 3; the arbitrage portfolios mostly beat the market, which indicates there is a
profit space when investors employ a contrarian strategy.

4.4 Short-term analysis

Table 4 shows the grand average excess cumulative returns of loser portfolio, winner
portfolio and arbitrage portfolio for short-term analysis (Formation period: 6 months, test
period: 3 months). The 3 months of testing periods give 22 tests for six overlapping years. In
comparison with market return, the winner portfolio returns keep slightly positive after
formation period which means the average of first 12 test periods winners are still winners
(compare with the market) and become negative since test period 13. Meanwhile, the loser
portfolio returns to keep a positive trend in the whole test period (except test 19 and 22) which
means the most test periods they beat the market. Moreover, the arbitrage portfolio return is

Grand average excess cumulative returns of portfolios

GAECR_W GAECR_L DG_L-W
Number of testing periods Mean t Mean t Mean t

Test 1 –0.0256 –0.5259 0.0118 0.3160 0.0374 0.8234

Test 2 –0.0736 –0.7810 –0.0220 –0.3695 0.0515 0.7062
Test 3 –0.0087 –0.4823 0.0169 0.2573 0.0256 0.3281
Test 4 –0.0079 –0.1961 –0.0176 –0.1914 –0.0097 –0.0752
Test 5 –0.0468 10.3763* 0.0205 0.1753 0.0674 0.5926
Test 6 –0.0989 3.6431* 0.0078 0.0458 0.1067 0.6263 Table 3.
Test 7 –0.1312 2.7278* 0.2280 10.6746* 0.3592 7.4999* Grand average excess
returns of winner
Test 8 –0.2072 4.0528* 0.3275 3.6703* 0.5348 3.8096* portfolios, loser
Note(s): GAECRW average excess cumulative returns of the winner portfolios after formation; GAECRL portfolios and
average excess cumulative returns of the loser portfolios after formation; DG_L-W Grand average excess arbitrage portfolios
cumulative returns of arbitrage portfolios (longing loser portfolios and shorting winner portfolios) after (testing period of
formation. *Significant at 5% level 6 months)
IJMF 0.6

0.5

0.4

0.3

0.2

0.1

0.0

Figure 3. -0.1
Grand average excess
cumulative returns of -0.2
winner portfolios, loser
portfolios and -0.3
arbitrage portfolios Test 1 Test 2 Test 3 Test 4 Test 5 Test 6 Test 7 Test 8
(testing period of
6 months)
GAECR_W GAECR_L DG_L-W

also positive except for test 19 and 22, which shows that the loser portfolios not only beat the
market but also are the winners. Figure 4 shows the graph of Table 4. In the 22 test periods,
the winner portfolio recovered to 0.08, which is the highest after formation period, it
outperforms the market but not the loser portfolio before test 13 and then even worse than the
market. Loser portfolios perform better than the market after formation period (except for test
19 and 22) with the highest return of 0.165 in an average of 18 tests periods.
With the comparison of all winner portfolios and loser portfolios in all formation periods
(1-year formation period), we find the first formation period (2009) winner portfolio has a
highest average excess return (2.673488) and the third formation (2011) period loser has the
lowest average excess return (0.66059). Based on this, we have chosen the first formation
period winner portfolio as the biggest winner portfolio and the third formation period loser
portfolio as the biggest loser portfolio (6-months formation period biggest winner: January to
June 2009; biggest loser: January to June 2011). We ranked the biggest winner and loser
portfolios six test periods before and after the formation period, and the results are reported in
Figures 5 and 6 below. The biggest loser portfolio performance better before and after the
formation period and winner perform worse before and after the formation period. The sharp
increase (decrease) of excess return is the overreaction of the “news” released in the market.

4.5 Additional tests

To further confirm the above results, we also applied regression analysis to test a return
reversal effect in a period of the market. Our regression model used is as follows:
AR1 ¼ α þ βAR0 þ ε:

Where 0 is the formation period, and 1 is the first test period after formation. The coefficient β
will be a measure of mean reversal. If there is a reversal, a negative β will be found in the
regression. And if there is a “big news” in the market, β should be significant, indicating the
abnormal return in the “news” time. Table 5 results show that the loser portfolios with
3-months test periods have significant negative coefficients, and sample loser portfolios of the
 Grand average excess cumulative returns of portfolios
GAECR_W GAECR_L DG_L-W
No. of testing period t-stat Mean Std. deviation t-stat Mean Std. deviation t-stat Mean Std. deviation

Test 1 0.9506 0.0191 0.06383 2.15* 0.086222 0.1268255 1.40 0.067032 0.150477
Test 2 0.07327 0.00103 0.04427 1.735 0.035569 0.0648255 1.395 0.034543 0.078281
Test 3 0.59651 0.00842 0.04462 2.218* 0.036875 0.0525785 1.241 0.028459 0.072518
Test 4 0.16343 0.00234 0.04533 2.469* 0.031068 0.0397908 1.713 0.028726 0.053023
Test 5 0.96050 0.00917 0.02863 3.769* 0.031849 0.0253480 1.727 0.022682 0.039406
Test 6 0.11781 0.00171 0.04367 4.856* 0.029006 0.0179217 1.760 0.027291 0.046532
Test 7 0.98205 0.01157 0.03331 4.086* 0.030368 0.0210239 1.167 0.018804 0.045575
Test 8 0.73492 0.00797 0.03066 4.882* 0.033213 0.0192436 2.028* 0.025248 0.035209
Test 9 1.66864 0.01839 0.02916 5.035* 0.029320 0.0154082 0.805 0.010933 0.035926
Test 10 1.52246 0.01674 0.02909 3.219* 0.025014 0.0205588 0.553 0.008277 0.039606
Test 11 1.73177 0.01643 0.02323 5.285* 0.031814 0.0147462 1.205 0.015389 0.031292
Test 12 0.33670 0.00533 0.03874 4.074* 0.033060 0.0198761 1.212 0.027735 0.056031
Test 13 0.09118 0.00180 0.04409 2.888* 0.034891 0.0270190 1.166 0.036689 0.070330
Test 14 0.32811 0.00682 0.04646 2.029* 0.030531 0.0336411 1.079 0.037349 0.077367
Test 15 0.65868 0.02159 0.06555 2.489* 0.046326 0.0372211 1.345 0.067915 0.101000
Test 16 0.64992 0.02361 0.07266 1.916* 0.056251 0.0587049 1.217 0.079865 0.131208
Test 17 0.77227 0.05336 0.11968 1.219 0.089564 0.1272427 1.003 0.142924 0.246782
Test 18 0.88555 0.09213 0.18020 1.026 0.165185 0.2787567 0.972 0.257320 0.458483
Test 19 0.19582 0.00317 0.02291 0.154 0.00070 0.0064560 0.333 0.003874 0.016457
Test 20 0.03630 0.00016 0.00614 1.084 0.015649 0.0204098 1.567 0.015807 0.014265
Test 21 0.02435 0.006823 0.031171
Test 22 0.08337 0.01240 0.09577
Note(s): GAECRW average excess cumulative returns of the winner portfolios after formation; GAECRL average excess cumulative returns of the loser portfolios after
formation; DG_L-W Grand average excess cumulative returns of arbitrage portfolios (longing loser portfolios and shorting winner portfolios) after formation. *Significant
at 5% level
Chinese stock
market
Overreaction

(testing period of
effect of

portfolios and
portfolios, loser
Table 4.
Grand average excess

3 months)
returns of winner

arbitrage portfolios
IJMF 0.30

0.25

0.20

0.15

0.10

0.05

0.00

Figure 4. -0.05
Grand average excess
cumulative returns of -0.10
winner portfolios, loser
S 1
S 2
S 3
S 4
S 5
S 6
S 7
8
ST _9
S 10
S 11
S 12
S 13
S 14
S 15
S 16
S 17
S 18
S 19
S 20
ST 21
2
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_

_2
portfolios and
TE _
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE T_
TE ST
S

arbitrage portfolios
TE

(testing period of
3 months)
GAECRW GAECRL DG

2.5

1.5

0.5
Figure 5. 0
Winner and loser -6 -4 -2 0 2 4 6
portfolios volatility -0.5
(ACER) test period of
6 months before and -1
after formation
Biggest Winner Biggest Losser

longer period also have negative coefficients but are not significant. Meanwhile, all winner
portfolios for both sample periods have insignificant positive coefficients. Based on these
results, the winner portfolios kept outperforming the market with lower abnormal returns in
test periods 1 and 2. They were still winners compared with the market; however, their
advantage was insignificant. The loser portfolios outperformed the market in test period 1
and turned back into losers in test period two. From the research sample period, the shorter
the formation and test periods, the more significant the reversal effect for both loser and
winner portfolios: there is a momentum effect in the 3 months after the formation period.
These results strongly show asymmetry between winner and loser portfolios.
Prior research on the overreaction hypotheses has tested long-run phenomena (3 years, 5
years) and short-run phenomena (1 week and 1 month). Medium-term tests were undertaken
by Kryzanowski and Zhang (1992) and Jegadeesh and Titman (1993), they reported different
 1.4 Overreaction
1.2 effect of
1 Chinese stock
0.8 market
0.6
0.4
0.2
0 Figure 6.
-8 -6 -4 -2 -0.2 0 2 4 6 8 Winner and loser
portfolios volatility
-0.4
(ACER) test period of
-0.6 3 months before and
after formation
Biggest winner Biggest Loser

Coefficients between formation period and test period

Formation 1 year, test 6 months Formation 6 months, test 3 months
Winner t-stat Winner t-stat

Test1 0.056056 1.537151 Test1 0.105313 1.633812

Test2 0.04019 0.450222 Test2 0.052856 1.228781

Loser Loser
Table 5.
Test1 0.07663 0.805752 Test1 0.27239 2.235732* Coefficients between
Test2 0.040088 0.38628 Test2 0.057375 0.795992 formation and test
Note(s): *Significant at the 0.05 level period

result, that is, return continuation, which supports the hypotheses of momentum rather than
a reversal. However, our results show that the reversal effect and overreaction exist, but are
insignificant in the test period, and shorter sample periods seem to have more explanatory
power. Therefore, we assume that the reversal effect and overreaction to news occurs in the
concise term after the formation period. Moreover, other issues need to be considered, that is,
time-period, stability, seasonality and size. We have considered each of them as given below.
Overreaction and subsequent reversal are a real phenomenon in the stock market and can
be expected across periods. In our test, the winner and loser portfolios in different periods show
statistically significant abnormal returns. The overreaction hypothesis is thus confirmed for
different periods. Seasonality has been introduced as the January effect and the Monday effect,
confirmed by Zarowin (1990) and Chopra et al. (1992) in the long-run, overreaction. However,
Bremer and Sweeney (1991) evidence for the January effect from the short-run perspective. Our
research is not influenced by monthly or weekly effects since its formation and test periods are
both over 3 months. The effect of size is ubiquitous in the study of overreaction and other
hypotheses about the stock market. As is known, size is correlated with a range of relevant
factors. However, size is not affecting these results as the sample includes firms from the SSE
50 index, which means that none in the sample were small or weak.

5. Conclusion
This study investigates the existence of short-term and intermediate-term overreaction in the
SSE Market. Using monthly data of samples with formation periods of 1 year and test periods
IJMF of 6 months, the loser portfolios of 50 sample stocks in the SSE 50 Index outperformed the
winner portfolios by 0.024 in 1 test period after formation and by a 0.141 average. In the
samples with formation periods of 6 months and test periods of 3 months, the loser portfolios
outperformed the winner portfolios by 0.025 in 1 test period after formation and by a 0.039
average. In this research, all stocks in the index represent respective industries and had high
capitalisation and liquidity. The arbitrage portfolios were formed with an approach minimum
cost and smooth execution. The findings of the study suggest that there are asymmetrical
overreactions in the stock market, especially for loser portfolios. The before-after test for the
biggest winner and loser portfolios shows that losers recovered and beat the market
immediately.
The findings of this study support the findings of Maheshwari and Dhankar (2015).
Tripathi and Agarwal, 2009, who worked on emerging markets. Contrary to the developed
markets where large institutional investors invest in stock markets, Chinese stock market
investors are inexperienced individuals who trade more frequently than foreign counterparts,
i.e. average stock holding period in China is two months (Zhang et al., 2018). According to
Nofsinger and Sias (1999), individual investors made more irrational decisions than
institutional investors. These investors’ irrational behaviour leads to high market volatility,
high transactions costs due to the short holding period and asymmetric information. The
results confirm the presence of high market volatility due to more individual investors than
institutional investors who lead to more irrational decisions making the confirm presence of
overreaction effect in Chinese stock market SSE index. The findings are also like Kang et al.
(2002) and Wu (2011) who explained the leading cause of abnormal return to being investor
overreaction to firm-specific information but this hold for the more short-term than for
intermediate-term like our results.
We have used different formation periods, and the results are strongly consistent for each.
There is no evidence that monthly returns and reversals are driven by seasonality or size. The
research implications of this study are for market behaviour researchers and academicians,
where they could find and reveal that how working under hypotheses of an overreaction;
gains can be made with cumulative abnormal return methodology, through arbitrage
portfolios construction. We also provide implications for investors with specific additional
support for the short and medium-term overreaction in the SSE for the period 2009–2014
using regression analysis. The winner portfolios kept outperforming the market with lower
abnormal returns in test periods 1 and 2 and were still winners compared with the market.
While the loser portfolios outperformed the market in test period 1 and turned back into losers
in the test period two. The study could also benefit government, policymakers and regulators
of the economy by studying how the presence of more individual investors than institutional
investors of China stock market leads to more irrational decisions giving rise to volatility. The
regulators could build favourable policies for institutional investors to give them the
incentive to invest more than individual investors through which market volatility could be
controlled.

References
Ahmad, Z. and Hussain, S. (2001), “KLSE long run overreaction and the Chinese new-year effect”,
Journal of Business Finance and Accounting, Vol. 28 Nos 1-2, pp. 63-105.
Ali, R., Ahmad, Z. and Anusakumar, S.V. (2011), “Stock market overreaction and trading volume:
evidence from Malaysia”, Asian Academy of Management Journal of Accounting and Finance,
Vol. 7 No. 2, pp. 103-119.
Alonso, A. and Rubio, G. (1990), “Overreaction in the Spanish equity market”, Journal of Banking and
Finance, Vol. 14 No. 2, pp. 469-481.
Antoniou, A., Galariotis, E.C. and Spyrou, S.I. (2005), “Contrarian profits and the overreaction Overreaction
hypothesis: the case of the Athens stock exchange”, European Financial Management, Vol. 11
No. 1, pp. 71-98. effect of
Atkins, A.B. and Dyl, E.A. (1990), “Price reversals, bid-ask spreads, and market efficiency”, Journal of
Chinese stock
Financial and Quantitative Analysis, Vol. 25 No. 4, pp. 535-547. market
Bailey, W. (1994), “Risk and return on China’s new stock markets: some preliminary evidence”, Pacific-
Basin Finance Journal, Vol. 2, pp. 243-60.
Balvers, R.J. and Wu, Y. (2006), “Momentum and mean reversion across national equity markets”,
Journal of Empirical Finance, Vol. 13 No. 1, pp. 24-48.
Balvers, R., Wu, Y. and Gilliland, E. (2000), “Mean reversion across national stock markets and
parametric contrarian investment strategies”, The Journal of Finance, Vol. 55 No. 2, pp. 745-772.
Banz, W. (1981), “The relationship between return and market value of common stock”, Journal of
Financial Economics, Vol. 9, pp. 3-18.
Basu, S. (1977), “Investment performance of common stocks in relation to their price-earnings
ratios: a test of the efficient market hypothesis”, The Journal of Finance, Vol. 32 No. 3,
pp. 663-682.
Baytas, A. and Cakici, N. (1999), “Do markets overreact: international evidence?”, Journal of Banking
and Finance, Vol. 23 No. 7, pp. 1121-1144.
Bekaert, G., Harvey, C.R. and Lundblad, C. (2007), “Liquidity and expected returns: lessons from
emerging markets”, Review of Financial Studies, Vol. 20 No. 6, pp. 1783-1831.
Bowman, R.G. and Iverson, D. (1998), “Short-run overreaction in the New Zealand stock market”,
Pacific-Basin Finance Journal, Vol. 6 No. 5, pp. 475-491.
Brailsford, T. (1992), “A test for the winner-loser anomaly in the Australian equity market: 1958–87”,
Journal of Business Finance and Accounting, Vol. 19 No. 2, pp. 225-241.
Bremer, M. and Sweeney, R.J. (1991), “The reversal of large stock-price decreases”, The Journal of
Finance, Vol. 46 No. 2, pp. 747-754.
Chan, K.C. (1988), “On the contrarian investment strategy”, Journal of Business, pp. 147-163.
Chopra, N., Lakonishok, J. and Ritter, J.R. (1992), “Measuring abnormal performance: do stocks
overreact?”, Journal of Financial Economics, Vol. 31 No. 2, pp. 235-268.
Clare, A. and Thomas, S. (1995), “The overreaction hypothesis and the UK stock market”, Journal of
Business Finance and Accounting, Vol. 22 No. 7, pp. 961-973.
Conrad, J. and Kaul, G. (1993), “Long-term market overreaction or biases in computed returns?”, The
Journal of Finance, Vol. 48 No. 1, pp. 39-63.
Cutler, D.M., Poterba, J.M. and Summers, L.H. (1991), “Speculative dynamics”, The Review of Economic
Studies, Vol. 58 No. 3, pp. 529-546.
Da Costa, N.C. (1994), “Overreaction in the Brazilian stock market”, Journal of Banking and Finance,
Vol. 18 No. 4, pp. 633-642.
Daniel, K., Hirshleifer, D. and Subrahmanyam, A. (1998), “Investor psychology and security market
under-and overreactions”, The Journal of Finance, Vol. 53 No. 6, pp. 1839-1885.
De Bondt, W.F. and Thaler, R. (1985), “Does the stock market overreact?”, The Journal of Finance,
Vol. 40 No. 3, pp. 793-805.
De Long, J.B., Shleifer, A., Summers, L.H. and Waldmann, R.J. (1990), “Noise trader risk in financial
markets”, Journal of Political Economy, pp. 703-738.
Dissanaike, G. (1994), “On the computation of returns in tests of the stock market overreaction
hypothesis”, Journal of Banking and Finance, Vol. 18 No. 6, pp. 1083-1094.
Dubois, M. and Bacmann, J.F. (1998), Contrarian Strategies and Cross-Autocorrelations in Stock
Returns: Evidence from France, Available at SSRN 138176.
IJMF Fama, E.F. and French, K.R. (1988), “Permanent and temporary components of stock prices”, Journal
of Political Economy, pp. 246-273.
Forner, C. and Marhuenda, J. (2004), “Momentum Returns in the Spanish Stock Market: Model
Misspecification or Investor Irrationality?”, EFMA 2004 Basel Meetings Paper.
Harvey, C.R. (1995), “The risk exposure of emerging equity markets”, The World Bank Economic
Review, Vol. 9 No. 1, pp. 19-50.
Hirshleifer, D. (2001), “Investor psychology and asset pricing”, The Journal of Finance, Vol. 56 No. 4,
pp. 1533-1597.
Jegadeesh, N. and Titman, S. (1993), “Returns to buying winners and selling losers: implications for
stock market efficiency”, The Journal of Finance, Vol. 48 No. 1, pp. 65-91.
Kang, J., Liu, M.H. and Ni, S.X. (2002), “Contrarian and momentum strategies in the China stock
market: 1993–2000”, Pacific-Basin Finance Journal, Vol. 10 No. 3, pp. 243-265.
Kryzanowski, L. and Zhang, H. (1992), “The contrarian investment strategy does not work in
Canadian markets”, Journal of Financial and Quantitative Analysis, Vol. 27 No. 3, pp. 383-395.
Lee, C.M. and Swaminathan, B. (2000), “Price momentum and trading volume”, The Journal of Finance,
Vol. 55 No. 5, pp. 2017-2069.
Lehmann, B. (1990), “Fads, martingales, and market efficiency”, Quarterly Journal of Economics,
Vol. 105, pp. 1-28.
Li, X., Xie, H., Chen, L., Wang, J. and Deng, X. (2014), “News Impact on stock price return via sentiment
analysis”, Knowledge-Based Systems, Vol. 69 No. 1, pp. 14-23, doi: 10.1016/j.knosys.2014.04.022.
Lim, K.P., Brooks, R.D. and Kim, J.H. (2008), “Financial crisis and stock market efficiency: empirical
evidence from Asian countries”, International Review of Financial Analysis, Vol. 17 No. 3,
pp. 571-591.
Lobe, S. and Rieks, J. (2011), “Short-term market overreaction on the frankfurt stock exchange”, The
Quarterly Review of Economics and Finance, Vol. 51 No. 2, pp. 113-123.
Locke, S. and Gupta, K. (2009), “Applicability of contrarian strategy in the Bombay stock exchange”,
Journal of Emerging Market Finance, Vol. 8 No. 2, pp. 165-189.
Ma, X. (1996), “Capital controls, market segmentation and stock prices: evidence from the Chinese
stock market”, Pacific-Basin Finance Journal, Vol. 4, pp. 219-39.
MacNeil, W.P. (2002), “You slay me: buffy as jurisprude of desire”, Cardozo Law Review, Vol. 24,
p. 2421.
Maheshwari, S. and Dhankar, R. (2015), “Long-run return reversal effect: a Re-examination in the
Indian stock market”, The Journal of Business Inquiry, Vol. 14 No. 2, pp. 59-78.
Malin, M. and Bornholt, G. (2013), “Long-term return reversal: evidence from international market
indices”, Journal of International Financial Markets, Institutions and Money, Vol. 25, pp. 1-17.
Naughton, T., Truong, C. and Veeraraghavan, M. (2008), “Momentum strategies and stock returns:
Chinese evidence”, Pacific-Basin Finance Journal, Vol. 16 No. 4, pp. 476-492.
Nofsinger, J. and Sias, R. (1999), “Herding and feedback trading by institutional and individual
investors”, J. Financ., Vol. 54 No. 6, pp. 2263-2295.
Poterba, J.M. and Summers, L.H. (1988), “Mean reversion in stock prices: evidence and implications”,
Journal of Financial Economics, Vol. 22 No. 1, pp. 27-59.
Reddy, K., Qamar, M.A.J. and Rao, M. (2019), “Return reversal effect in Shanghai A share market”,
Managerial Finance, Vol. 45 No. 6, pp. 698-715.
Richards, A.J. (1997), “Winner-loser reversals in national stock market indices: can they be explained?”,
The Journal of Finance, Vol. 52 No. 5, pp. 2129-2144.
Rosenberg, B., Reid, K. and Lanstein, R. (1985), “Persuasive evidence of market inefficiency”, Journal
of Portfolio Management, Vol. 11 No. 3, pp. 9-16.
Schoenfeld, S.A. (2011), Active Index Investing: Maximizing Portfolio Performance and Minimizing Overreaction
Risk through Global Index Strategies, Vol. 138, John Wiley & Sons.
effect of
Sharma, S.S., Narayan, P. and Thuraisamy, K. (2015), “Time-varying herding behavior, global
financial crisis, and the Chinese stock market”, Review of Pacific Basin Financial Markets and
Chinese stock
Policies, Vol. 18 No. 2, p. 1550009. market
Shumway, T. and Wu, G. (2006), “Does disposition drive momentum?”, University of Michigan
working paper.
Stein, J. (1989), “Overreactions in the options market”, The Journal of Finance, Vol. 44 No. 4,
pp. 1011-1023.
Stock, D. (1990), “Winner and loser anomalies in the German stock market”, Journal of Institutional
and Theoretical Economics (JITE)/Zeitschrift f€
ur die gesamte Staatswissenschaft, pp. 518-529.
Su, D. and Fleisher, B. (1998), “Risk, return and regulation in Chinese stock markets”, Journal of
Economics and Business, Vol. 50, pp. 239-56.
Susan, S.S., Paresh, K.N. and Kannan, S.T. (2015), “Time-varying herding behavior, global financial
crisis, and the Chinese stock market”, Review of Pacific Basin Financial Markets and Policies,
Vol. 18 No. 2, p. 1550009, doi: 10.1142/S0219091515500095.
Swallow, S. and Fox, M.A. (1998), Long Run Overreaction on the New Zealand Stock Exchange, Lincoln
University, Commerce Division.
Tripathi, V. and Gupta, S. (2009), “Overreaction effect in Indian stock market”, Asian Journal of
Business and Accounting, pp. 1-2.
Wang, C. and Chin, S. (2004), “Profitability of return and volume-based investment strategies in
China’s stock market”, Pacific-Basin Finance Journal, Vol. 12 No. 5, pp. 541-564.
Wang, J., Burton, B.M. and Power, D.M. (2004), “Analysis of the overreaction effect in the Chinese
stock market”, Applied Economics Letters, Vol. 11 No. 7, pp. 437-442.
Wu, Y. (2011), “Momentum trading, mean reversal and overreaction in Chinese stock market”, Review
of Quantitative Finance and Accounting, Vol. 37 No. 3, pp. 301-323.
Zarowin, P. (1990), “Size, seasonality, and stock market overreaction”, Journal of Financial and
Quantitative Analysis, Vol. 25 No. 1, pp. 113-125.
Zhang, W., Wang, G.Y., Wang, X.C., Xiong, X. and Lei, X. (2018), “Profitability of reversal strategies: a
modified version of the Carhart model in China”, Economic Modelling, Vol. 69, pp. 26-37.
Zhao, Y., Wei, J., Li, H., Yan, Y., Zhou, W., Yu, D. and Zhao, Q. (2016), “A polymer scaffold for self-
healing perovskite solar cells”, Nature Communications, Vol. 7, p. 10228.
Zijun, S. and Lin, H. (2015), “A neuron model with dendritic nonlinearity for predicting the influence of
overreaction in Shanghai stock market”, Chinese Studies, Vol. 4, pp. 1-9.

Further reading
Dhankar, R.S. and Maheshwari, S. (2019), “Stock markets overreaction”, in Capital Markets and
Investment Decision Making, Springer, New Delhi, pp. 83-107.
Fama, E.F. (1970), “Efficient capital markets: a review of theory and empirical work”, The Journal of
Finance, Vol. 25 No. 2, pp. 383-417.
Farag, H. (2015), “The influence of price limits on overreaction in emerging markets: evidence from the
Egyptian stock market”, The Quarterly Review of Economics and Finance, Vol. 58, pp. 190-199,
doi: 10.1016/j.qref.2015.01.003.
Fung, A.K.W. (2000), “Overreaction in the Hong Kong stock market”, Global Finance Journal, Vol. 10
No. 2, pp. 223-230.
Hong, H. and Stein, J.C. (1999), “A unified theory of underreaction, momentum trading, and
overreaction in -aasset -markets”, The Journal of Finance, Vol. 54 No. 6, pp. 2143-2184.
IJMF Maher, D. and Parikh, A. (2011), “Short-term under/overreaction, anticipation or uncertainty
avoidance? Evidence from India”, Journal of International Financial Markets, Institutions and
Money, Vol. 21 No. 4, pp. 560-584.
Morse, P.B., Black, F., Jarrow, R., Elton, E. and Watts, R. (1985), “Does the stock market overreact?
Werner Fm de Bondt and richard thaler”, The Journal of Finance, Vol. 40 No. 3.
Phan, D.H.B., Sharma, S.S. and Narayan, P.K. (2015), “Stock return forecasting: some new evidence”,
International Review of Financial Analysis, Vol. 40, pp. 38-51.
Wang, C. (2004), “Relative strength strategies in China’s stock market: 1994–2000”, Pacific-Basin
Finance Journal, Vol. 12 No. 2, pp. 159-177.
Wang, Y., Gunasekarage, A. and Power, D.M. (2005), “Return and volatility spillovers from developed
to emerging capital markets: the case of South Asia”, Contemporary Studies in Economic and
Financial Analysis, Vol. 86, pp. 139-166.
Wang, Y., Liu, L. and Gu, R. (2009), “Analysis of efficiency for Shenzhen stock market based on
multifractal detrended fluctuation analysis”, International Review of Financial Analysis, Vol. 18
No. 5, pp. 271-276.

Corresponding author
Muhammad Ali Jibran Qamar can be contacted at: majqamar@gmail.com

For instructions on how to order reprints of this article, please visit our website:
www.emeraldgrouppublishing.com/licensing/reprints.htm
Or contact us for further details: permissions@emeraldinsight.com

View publication stats