Is There a Momentum Effect in Chinese STAR MARKET? An
Empirical Analysis based on 223 Stocks in 2019-2021
Yiwen Liu
1,a
and Yuze Jiang
2,b
1
Guangdong University of Technology, Guangzhou, China
2
Monash University, Melbourne, Australia
Keywords: SSE STAR MARKET, Python, Turnover Rate, Momentum Effect.
Abstract: With the deepening of the reform and opening-up process, China's capital market has gradually become
mature. Established in 2019, the SSE STAR MARKET (written below as SSE) is a special board in China
specifically for high-tech enterprises to solve the problem of the high threshold for listing on the mainboard,
and to promote the development of innovative enterprises. The characteristics of the SSE are quite different
from those of the mainboard, so this paper use Python with weekly trading data for 223 Stocks in SSE to
investigate whether there is a momentum effect in the SSE by constructing a winner 's portfolio and a loser's
portfolio for the first time. What’s more, this paper investigates the relationship between the turnover rate and
the momentum effect. The study finds that there is no momentum effect in the SSE STAR MARKET in the
short term. However, there is a momentum effect within the long term and the extent of the momentum effect
deepens in the market with low monthly turnover rates.
1 INTRODUCTION
The momentum effect refers to the recent good
performance of monetary assets in the forthcoming
period will also maintain its better performance. In
some mature securities markets, investors have
started to buy and hold the financial assets of the
winning portfolio for a period of time, while selling
the financial assets of the losing portfolio to achieve
the momentum effect of investment strategy. In recent
years, domestic investors have also begun to look for
traces of the momentum effect in China's stock
market.
SSE, one of the components of China's stock
market, was established in July 2019, and it promotes
the process of technology-based and innovative
enterprises, and solves their concentration of
financing problems. As a place of innovation and
reform in the progress of China's capital market, the
SSE has promoted the development of Chinese
innovation into a golden age and brought historic
development opportunities for relevant high-tech
enterprises. At the same time, most of the listed
companies in the SSE have characteristics of high
potential, high return and high risk, and more and
more investors are focusing their attention on this
area and naturally thinking about whether they can
construct momentum effect strategies in the SSE to
gain the excess returns. The existing literature mainly
focuses on the analysis of the A-share market as a
whole, mostly using monthly and annual frequency
data, while the findings would not be the same if the
results were directly covered for the whole A-share
market due to the different sample subjects and
periods. Moreover, there are just a few papers that
refine the Chinese stock market and study the
momentum effect of a particular sector separately.
Considering that the price elasticity of the board is
greater than that of the mainboard, this paper adopts
the weekly data related to listed companies in the SSE
during 2019-2021 for empirical analysis to examine
the existence of the momentum effect and explore the
influencing factors for the first time, providing
investment strategy suggestions and references for
relevant investors.
2 LITERATURE REVIEW
2.1 Existence of Momentum Effects in
The Stock Market
The momentum effect has been examined and
analyzed by scholars around the world in both
66
Liu, Y. and Jiang, Y.
Is There a Momentum Effect in Chinese STAR MARKET? An Empirical Analysis based on 223 Stocks in 2019-2021.
DOI: 10.5220/0011723400003607
In Proceedings of the 1st International Conference on Public Management, Digital Economy and Internet Technology (ICPDI 2022), pages 66-71
ISBN: 978-989-758-620-0
Copyright
c
2023 by SCITEPRESS Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
established and emerging stock markets to provide
evidence for its existence. In foreign research,
Jegadeesh & Titman (Jegadeesh, 1993), the
originators of the momentum effect study, concluded
that stock returns are short-run persistent by studying
daily frequency data of individual stocks in the U.S.
stock market from 1965 to 1989. Later Fama &
French (Fama, 2012) on stock markets in the Asia-
Pacific region confirmed the existence of the
momentum effect to be different.
On the domestic side, the researches on the
momentum effect have been considered in different
directions due to the complexity of the A-share
market. Firstly, in terms of whether there is a short-
term momentum effect in the A-share market, the
conclusions of domestic scholars are more uniform,
led by Gao Qiuming, Hu Conghui, and Yan Xiang
(Gao, 2014), and Song Guanghui, Dong Yongqi, and
Chen Yang Yang (Song, 2017), who use weekly
frequency data and conduct a study based on the
overlap method, conclude that there is a significant
short-term (within 1 month) momentum effect in the
A-share market.
2.2 Influencing Factors of Momentum
Effects in The Domestic Stock
Market
Most of the analyses of the factors influencing the
momentum effect in Chinese stock markets have
mainly focused on two points. Firstly, the impact of
the domestic short selling mechanism: Qiuming Gao,
Conghui Hu, and Xiang Yan (Gao, 2014)
find that
restrictions on short selling promote short-term
momentum effects; Tianhui Zheng (Zheng, 2017)
distinguishes whether the underlying can be financed
and financed and finds that the underlying stocks that
can be financed and financed have more significant
momentum effects than those that cannot be financed
and financed. Secondly, the effect of turnover rate: Li,
Jiangping (Li, 2020) first used daily frequency data
of Land Stock Exchange from 2017 to 2019 and
concluded that Land Stock Exchange still has
significant momentum effect in the case of low
turnover rate.
3 RESEARCH HYPOTHESIS
It has been practically two years since the start-up of
the SSE, experiencing impacts such as the epidemic
and changes in the international political environment
on the way. The SSE showed an overall trend of high
growth and achieve dazzling results. This
undoubtedly reflects the huge potential and upside of
SSE and attracts a large number of investors to buy
into it, causing the share prices of these listed
companies to rise in a longer time, thus easily leading
to the momentum effect in a long-term perspective.
However, the investor structure of SSE is special,
mainly institutional investors and investors with
superior professional levels. The long holding time of
such investors solves to a certain extent the problem
of massive selling when encountering a bear market,
which leads to a sharp fall in stock prices and
stabilizes the market price, making it difficult to form
a momentum effect in the short term. In the actual
market environment, there is also the same herding
effect that retail investors are prone to form at the
initial stage of buying in the SSE, i.e., the stock price
climbs up one after another. Based on the above
analysis, the first hypothesis of this paper is proposed.
Hypothesis 1: There is no short-term momentum
effect but a long-term momentum effect for SSE
stocks.
As mentioned earlier in this paper, most domestic
scholars revolve around the mechanism of our stock
(difficult to short) and turnover rate. As a measure of
stock liquidity, low turnover rate responds to low
liquidity, and according to the explanation for the
creation of momentum effect in behavioral finance:
the lack of response from investors causes, which
means that the information dissemination in the
market is slower, easily causing information bias and
asymmetry, resulting in stock price deviation from its
own true value. Therefore, based on the finding of
scholar Li Jiangping (Li, 2020) that Land Stock
Exchange long position stocks have momentum
effect despite low turnover rate, this paper explores
whether the turnover rate affects the existence of
momentum effect in SSE and proposes the second
hypothesis of this paper.
Hypothesis 2: With the low turnover rate, stocks
in the SSE still have a significant medium-term
momentum effect.
Is There a Momentum Effect in Chinese STAR MARKET? An Empirical Analysis based on 223 Stocks in 2019-2021
67
Figure 1: Momentum effect strategy construction method.
4 RESEARCH METHODOLOGY
AND DATA SOURCES
4.1
Research Methodology
The model used throughout this paper is the
overlapping sampling method used by Jegadeesh &
Titman (Jegadeesh, 1993), finding whether there is a
momentum effect in the SSE. In this paper, the period
is divided into two periods, the first period A is the
observation period and the second period B is the
holding period. The holding period of 1-2 weeks is
defined as short term, 4 weeks as medium term, and
8 weeks as long term. Also, the time lengths of A and
B are taken as 1 week, 2 weeks, 4 weeks and 8 weeks
respectively, i.e., 16 different groups of portfolios are
paired, and the stocks with the top 10% returns in the
observation period are recorded as winners'
portfolios, while the bottom 10% stocks are recorded
as losers' portfolios. The stocks in each group are
bought according to the same weight, and then the
average return of the winner portfolio and the loser
portfolio are calculated separately during the holding
period, and finally, the difference between the winner
portfolio and the loser portfolio (W-L) is calculated.
Also, a one-week space between the observation
period and the holding period is used to prevent the
occurrence of the overrun-lag effect, which becomes
the form of (A, 1, B) strategy.
In addition, in order to study the effect of turnover
rate on the emergence of momentum effect of SSE
stocks, this paper mimics the overlapping sampling
method model by this following: First, the turnover
rate period C, i.e. (C, A, B), is set before the
observation period and the position period,
considering the short length of SSE establishment,
which will result in too small a study sample, so no
space-time is set here. The first 30% of stocks are the
high turnover rate portfolio, and the last 30% are the
low turnover rate portfolio. Then in the low turnover
portfolio in the observation period for the returned
ranking from the largest to the smallest, take the top
10% return stock portfolio as the winner portfolio, on
the contrary, the lowest 10% return stock portfolio as
the loser portfolio. Finally, the difference in return
between winner and loser portfolio in the holding
period is measured. In this paper, the value of C is
fixed for one month, and A and B are taken for 1
week, 2 weeks and 4 weeks, respectively, to construct
9 groups of strategies to judge the impact of turnover
rate.
4.2 Data Sources
This paper uses the Choice financial data platform of
Oriental Wealth to obtain the return and monthly
turnover data of the stocks of the CoC, with the
sample time range from October 2019 to May 2021,
i.e., starting from the second month of the inception
of the CoC. In this paper, we analyze the existence of
the momentum effect of the Cochrane Board in terms
of weekly time. Each strategy has a fixed observation
period and holding period. There are 16 portfolio
strategies, stocks with missing data and historical
returns of less than 1 month removed. Depending on
the above data criteria, a total of 223 stocks of SSE is
derived for research and analysis in this paper.
In this paper, the return calculation of the winner
portfolio, the loser portfolio and the difference
between the two are done through the WindQuant
platform by introducing the WindPy for python
quantitative back testing, and calculating the return of
the stock portfolio under different parameter settings
respectively. Using a strategy with a 1-week
observation period and a 1-week holding period (1, 1,
1) as an example, Figure 2 shows the average yield of
the winner's portfolio, the loser's portfolio and the
difference between the two over different periods. (If
you need the code of this paper, please feel free to
contact the author by email).
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
68
Figure 2: (1, 1, 1) Strategy grouping returns.
Table 1: Returns of different strategy combinations of SSE.
4.3
Description of Variables and
Definition of Indicators
The average return of the winner's portfolio over each
holding period is:
𝑅
Σ
.
𝛱
1 r

1
0.1xN
1
The average return of the loser portfolio over each
holding period is
𝑅
Σ
.
𝛱
1 r

1
0.1xN
2
The cumulative average return of the portfolio
over the sample time period is:
𝐶𝑅
/
Σ
𝑅
/
Y
3
Difference between winner's and loser's
portfolios:
𝑆𝑝𝑟𝑒𝑎𝑑

∑
𝛱
1 𝑟

1
0.1𝑥𝑁
𝛱
1 𝑟

1
0.1𝑥𝑁
4
5 EMPIRICAL RESULTS AND
ANALYSIS
5.1 Exploring the Existence of
Momentum Effects under Different
Strategies in the SSE Market
In Table 1 below, the average return results of the
winning portfolio, the losing portfolio, and the
portfolio difference under 16 different observation
and holding periods are presented, and a one-sample
t-test is conducted on the portfolio difference, i.e., the
zero-cost momentum portfolio, for each set of
strategies, with the original hypothesis that the mean
value of the return on the difference portfolio is zero,
as a test of whether the momentum effect is
significant. A positive return means that the
momentum effect exists, while the opposite is not
true. The following information can be consulted on
the table.
As shown in Table 1 (*** p<0.01, ** p<0.05, *
Is There a Momentum Effect in Chinese STAR MARKET? An Empirical Analysis based on 223 Stocks in 2019-2021
69
p<0.1), the maximum return on portfolio spread is
22.04% for the (4,2) strategy and the minimum return
is -64.98% for the (2,2) strategy. In these 16
strategies, the (1-1), (1-8), (2-8), (4-1), (4-2), (4-4),
and (4-8) spreads perform positively, most of which
are concentrated in the medium to long-term holding
period and the observation period of 4 weeks. To
address the issue of significance, the T-statistics of the
above portfolios with positive returns on difference is
significant at the 90% confidence level for strategies
(2, 8) and (4, 8), and at the 95% confidence level for
strategies (1, 8), while strategies (2, 2), (8, 1), (8, 2)
and (8, 4) show insignificant momentum effects.
When the observation period is 1 to 2 weeks, only the
(1, 1) portfolio shows a momentum effect and both
the winner portfolio and the loser portfolio have
negative average returns, i.e., the momentum effect
arises from the lower returns of the loser portfolio;
when the observation period is 4 weeks, the reason
comes from the high returns of the winner portfolio
and the low returns of the loser portfolio, indicating
that the stock rally in the SSE has some sustainability.
It is worth noting that when the observation period is
8 weeks, the portfolio spreads are all negative, and
after a longer period of better performance, there is a
significant pullback during the holding period,
indicating a probability of "momentum collapse" in
the SSE market.
The findings of this paper on whether there is a
momentum effect in the SSE market are different
from those of other scholars on China's A-share
market but are more consistent with the findings of
scholars on SECOND BOARD. In summary,
hypothesis 1 is correct that there is no momentum
effect for stocks in the SSE in the short term (1-2
weeks holding period) and medium term (4 weeks
holding period), but there is a momentum effect for
the long term (8 weeks holding period).
5.2
Exploring the Effect of Monthly
Turnover rate on the Momentum
Effect in the SSE Market
In order to investigate whether there is an effect of
turnover rate on the momentum effect in the SSE
market, this paper presents the average return results
of winning portfolios, losing portfolios, and portfolio
spreads under the influence of monthly turnover rate
for nine different observation and holding periods in
Table 2 below, and performs a one-sample t-test on
the portfolio spreads for each group of strategies with
the following results.
As shown in Table 2, among the three medium-
term strategy portfolios with a four-week holding
period, the return on the difference between the (1, 4)
and (4, 4) long-term portfolios is significantly
positive at the 90% confidence level, but unlike the
normal situation, the short-term strategy portfolios
also show a momentum effect under the low turnover
condition, with positive returns on the portfolios (1,
1), (1, 2), (2, 1), (2, 2), and (4, 1). And the t-statistics
of (1, 2), (2, 1), (2, 2) are significant at 90%
confidence level. This is a good indication that the
SSE market has a significant medium-term
momentum effect despite the low turnover rate and a
short-term momentum effect that do not exist in the
normal situation. The reason for this phenomenon
Table 2: Returns of different strategies in SSE under low turnover rate.
ICPDI 2022 - International Conference on Public Management, Digital Economy and Internet Technology
70
maybe because the low turnover rate implies that the
information transmission rate in the SSE market is
slow, investors' response is insufficient, and stock
prices cannot return to their intrinsic value level
quickly enough to eliminate the noise problem in a
short period, thus leading to a short-term momentum
effect in the SSE market. In summary, hypothesis two
holds and stocks in the SSE still have significant
medium-term momentum effects in the presence of
low turnover rates.
6 CONCLUSIONS AND
RECOMMENDATIONS
In this paper, we study 233 stocks in the period from
October 2019 to May 2021 in the SSE, finding that
there is no momentum effect in the SSE market in the
short (1-2 weeks) and medium term (4 weeks), and a
significant momentum effect in the long term (8
weeks); using low turnover as a precondition, the SSE
market shows momentum effects in the short,
medium, and long term. The findings in this paper are
different from those of the mainboard of China,
which are related to the differences in the listed
companies, investor structure and policies of the
board. This paper also demonstrates that the market
is still not efficient in China. Therefore, this paper
suggests the following recommendations to both
regulators and the investors.
For the regulators, it is important to improve the
information disclosure system and increase the
transparency of trading in the SSE market. It is the
existence of asymmetric information and slow
delivery that leads to the momentum effect.
Therefore, regulators should supervise the timeliness
of information disclosure of listed companies and
establish mechanisms to ensure the authenticity and
validity of the information provided by listed
companies.
For the investors, to strengthen investment
learning, reduce excessive speculation and other
irrational behavior. China's retail investors are not
well educated in investment, easy to form a herd
effect and follow the herd behavior, affected by the
noise information. Investors should look at the
market volatility rationally, do not blindly select
stocks, while building a portfolio of momentum
strategies under different circumstances, such as the
results of the empirical analysis of this paper, in the
case of low turnover rate, good risk management, and
in line with their investment plan, constitute a
portfolio of momentum strategies to obtain a market
excess return.
REFERENCES
Fama E F, French K R. Size, value, and momentum in
international stock returns [J]. Journal of Financial
Economics, 2012, 105(3): 457 -472.
Gao Qiuming, Hu Conghui, Yan Xiang. On Characteristics
and Formation Mechanisms of Momentum Effect in
China’s A—share Market [J]. Journal of Finance and
Economics, 2014,40(02):97-107.
Jegadeesh, Narasimhan, and Sheridan Titman. Returns to
buying winners and selling losers: implications for
stock market efficiency [J]. Journal of Finance, 1993,
48: 65-91.
Li Jiangping. An empirical study on the momentum effect
of Shanghai-Hong Kong Stock Connect program long
position stocks [J]. Shanghai Finance, 2020(04):22-30.
Qiang Hong. Progress of research on momentum effects in
China [J]. Review of Economic Research, 2018(50):20-
30.
Song Guanghui, Dong Yongqi, Chen Yang Yang, and Xu
Lin, Liquidity and the Momentum Effect of China's
Stock Market: In -depth Study Based on Fama -French
Five Factors Model [J]. Financial Economics Research,
Vol. 1, 2017
Tianhui Zheng, Research on the effect of heterogeneous
beliefs on momentum effect under short selling
mechanism, Master's thesis, Harbin Institute of
Technology, 2017
Is There a Momentum Effect in Chinese STAR MARKET? An Empirical Analysis based on 223 Stocks in 2019-2021
71