Available via license: CC BY 4.0
Content may be subject to copyright.
Studying the Impact of Media Coverage on the Stock Market
Using Mediating Effects
Zhang Ziyue[0009-0000-4503-6128] 1,a,*
1Sichuan International Studies University
a. 3452593485@qq.com
*corresponding author
Abstract: As a publisher of information on capital market transactions, media coverage has
become a hot topic of research for modern scholars. In this paper, stocks with trading data
within 2013.1.1-2020.1.1 on the SSE and SZSE are used as research samples, and the number
of media reports, investor sentiment, and stock returns are used as core variables, to
investigate the transmission relationship between these three things by using a mediating
effect model. The relationship between the three variables is reported systematically and
directly. This study concludes that investor sentiment plays a partially mediating role in the
impact of media attention on the stock market. Investor sentiment as a medium of media
information translates perceptions into behavior and thus influences investors' investment
decisions. The news gives investors some informational advantage reduced the level of
information asymmetry and has the most pronounced impact on investors, media involvement
diminishes the role of investor sentiment on stock investments.
Keywords: media attention, investor sentiment, stock reporting rate, information asymmetry.
1. Introduction
The role of the media is becoming increasingly non-negotiable with the rapid development of the
internet and is particularly important in the capital markets. As the main channel for investors to
receive information, the media plays a key role in the stock market by disclosing information about
listed companies and exposing scandals or other factors that influence investors' decisions.
On July 20, 2015, the China Securities Regulatory Commission (CSRC) issued an urgent
disinformation notice, stating that the report of Caijing that "the CSRC is studying the exit plan of
stabilization funds is untrue". The CSRC spokesperson indicated that it is irresponsible for the media
to report on the market by events with significant impact without verifying with the regulator and that
the CSRC will continue to complete the work related to stabilizing the market in the next phase. After
the SEC clarifying lagged, the form of the market index decline reversed and rose sharply. This
incident illustrates that media coverage and evaluation of state sector decisions guide investors'
investment choices on the stock market.
This paper investigates the transmission mechanism of media coverage on stock market changes
by using the number of news reports as a proxy variable for media attention, stock turnover rate as a
proxy variable for investor sentiment, and stock return rate as a proxy variable for stock market
changes, and investor sentiment as a mediating variable. The existing literature has more often studied
any two of the three variables or studied the impact of media on IPO pricing either that or media
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
© 2023 The Authors. This is an open access article distributed under the terms of the Creative Commons Attribution License 4.0
(https://creativecommons.org/licenses/by/4.0/).
118
sentiment's influence on investor behavior, etc. Few systematic and direct studies have been
conducted on all three. In this paper, we discuss them two by two separately discussions and then
analyze the three variables together. So that readers can have a better understanding of how much
mediating effect investor sentiment plays as a mediating variable. How does it play a mediating effect?
The structure of this paper is arranged: Chapter 1 presents the research questions and the
innovation points of the paper based on the research background. Chapter 2 is a literature review,
based on the efficient market theory, behavioral finance theory, and information asymmetry theory,
and collects relevant literature from two perspectives respectively, a literature review on media
coverage and investor sentiment and a literature review on investor sentiment and the stock market.
Chapter 3 is the theoretical analysis and research hypothesis, which proposes three hypotheses based
on the research questions the effect of media attention on investor sentiment, the effect of investor
sentiment on stock market changes, and the mediating role of investor sentiment. Chapter 4 is the
introduction of the data and data definitions, which first proposes to use stocks with trading data from
2013.1.1 to 2020.1.1 on SSE and SZSE as the research sample. In Chapter 5, descriptive statistical
analysis and mediated effects regression analysis are performed, and finally, the regression results are
analyzed. Chapter 6 gives a conclusion based on the research.
2. Literature Review
2.1. A literature review on media coverage and investor sentiment
The media is a branch independent of politics and law and is self-contained, but it has an important
influence on the financial market. With the development of the internet, the role of media in reporting
public information and transmitting sentiment has become more and more prominent. It also has
become a research hotspot for domestic and foreign scholars. In EMH, media coverage changes the
information structure of investors to further influence the stock market. But in behavioral finance,
media coverage affects the stock market by influencing investors' behavior, because investors are the
medium through which media coverage acts on the stock market. News about fundamental stock
information can have a long-term impact, and investors react after the neutral publication of
information by the media, and this reaction time causes the media not to act on stocks immediately.
However, if there is too much emotional element in media coverage, it can lead to a synchronized
change in stocks immediately after a shock of news coverage, but also a reversal phenomenon
afterward [1],[2]. These reactions create volatility in investor sentiment and trading markets,
respectively [3].
Investors in the stock market are usually divided into two categories: institutional investors and
individual investors. According to the China Securities Investor Protection Foundation (SIPF) data
as of the end of 2020, the number of retail investors in China reached 320 million, so individual
investors account for a relatively large share of the country. News information is the most widely
available channel for them to obtain external information, but the behavior of such investors tends to
aggregate, and irrational behavior of individuals hit by news may lead other investors and lead to a
herding effect. However, more information leads to more opportunities for profitable trading when
media articles make information public leading to a decrease in the frequency of insider trading and
a decrease in information asymmetry between managers and investors [4]. The role of media coverage
is reflected in providing information value mainly in the primary market, which reduces the degree
of information asymmetry, and providing sentiment value in the secondary market, which promotes
irrational sentiment and reduces the efficiency of asset pricing [5]. Soon-Ho Kim Dongcheol Kim [6]
collect 32 million data on yahoo message boards about financial manned companies, and find that
when investor sentiment is high, retail asset allocation demand expands, making prices higher than
asset values, but prices also return to fundamental values when investor sentiment recovers.
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
119
In addition, some researchers classify media sentiment based on salient tone words in media
coverage Hanna, A. J., Turner, J. D. and Walker, C. B. [7] quantifying investor sentiment using
Financial Times tone words as an indicator of investor sentiment finds that negative sentiment is
typically smaller in bull markets and that the standard deviation of pessimism is inversely
proportional to investment returns. Engelberg [8] models media information and media sentiment,
and then classifies investor behavior and news coverage by region, with local media coverage
tendencies correlated with local stock trading volume. However, instead of focusing on media
sentiment, this paper uses the number of media reports as a proxy variable for social concern.
2.2. A literature review on investor sentiment and stock markets
In the classical financial theory, investors were perfectly rational and mistakes could cancel each
other out, and investor sentiment did not play a role in realized and expected stock returns. Since
some anomalous effects in the stock market cannot be explained by EMH, behavioral finance has
been increasingly promoted. Behavioral finance supports the idea that there is an interaction between
investor sentiment and stock volatility [9]. When there is negative or positive news in the market, the
resulting investor sentiment tends to lead to the simultaneous bullish or bearish sentiment among
other investors in the market, who then make simultaneous buying or selling transactions, and stock
prices will thus produce abnormal fluctuations, which may in turn cause changes in investor sentiment.
Nowadays, investor sentiment affects the stock market and has become a consensus among some
scholars. Sayim [10] proved through research that stock index returns are lower when investor
sentiment is high, while there is no significant effect when investor sentiment is low. Rishad [11]
finds that irrational investor sentiment leads to asymmetric excessive stock market volatility.
Baker and Wurgler [12] confirm that broad-based investor sentiment has a cross-sectional effect,
with less preference for more volatile and riskier stocks when investor sentiment is high but reverses
when investor sentiment is low. Baek [13] uses a composite sentiment maker indicator to examine
the relationship between investor sentiment and stock returns, prices, and dividends. They find that
investor sentiment and stock returns are closely related and are more pronounced under extreme
markets. This effect is significantly positive, and the higher the investor sentiment, the more
pronounced the stock market volatility and negatively related to the market interest rate [14].
However, scholars have inconsistent views on the predictive effect of investor sentiment on the
stock market. Chung, Mei-Ling[15] and Yin, H. [16] argue that investor sentiment had a better
predictive effect on future stock market volatility, that is the lagged effect of investor sentiment is
stronger. Maik Schmeling [17] confirms that investor sentiment and stock reporting are correlated in
an international context and that sentiment predicts stock market reporting negatively. However,
Soon-Ho Kim [6] argues against the predictive effect of investor sentiment on future stock returns,
volatility, and trading volume.
3. Theoretical analysis and formulation of hypotheses
In efficient markets, investors are rational and it is impossible to obtain excess returns by predicting
stock prices. Fama[18] classifies efficient markets into weak efficient, semi-strong efficient, and
strong efficient forms. Returns are obtained through all past information in weak EMH, all public
information in semi-strong EMH, and all public and private information in strong EMH. The
possibility of gaining excess returns under each type of market is almost zero. As EMH gradually
fails to account for anomalies such as calendar effect, scale effect, and equity premium, more scholars
find that behavioral finance can provide a more reasonable explanation for investor behavior, and
traditional finance is thus greatly impacted, and the media is gaining importance as a distributor of
information in the market.
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
120
Behavioral finance argues that markets are not fully efficient, which is expressed through irrational
investors' emotions and limited arbitrage. Irrational investor sentiment and limited information cause
pessimistic or optimistic judgments about the future value of stocks, while limited arbitrage is caused
by imperfect market mechanisms.
There is information asymmetry in irrational markets, or information asymmetry is one of the
causes of investor sentiment. Akerlof [19] proposed the theory of information asymmetry, which
rejects the assumption that investors have full access to public information in the market. In
behavioral finance theory, media coverage affects investor sentiment, which in turn affects investor
behavior and finally capital markets. The emergence of media coverage has, to a certain extent,
weakened the degree of information asymmetry between managers and investors, weakened the
agency problem, and to a certain extent, compensated for the information deficit of individual
investors when trading in the capital market and reduced investors' uncertainty about the company.
However, due to the limited attention span of rational investors, the media can direct the attention of
investors on the one hand, and on the other hand, they convey media sentiment to investors, which
influences their judgment and causes volatility in the stock market. Maslyuk-Escobedo Svetlana[20]
by studying the impact of the U.S. media sentiment index on energy market returns concludes that
futures and spot returns of energy commodities fluctuate along with the correlation, but large
sentiment-driven investment behavior with high investor sentiment may lead to overvalued stock
prices, causing an overreaction to push up stock prices in the short term. Based on the above research
and theory, this paper proposes that.
H1: Media attention is positively correlated with investor sentiment
H2: There is a significant effect of investor sentiment on stock market returns
Combining the two relational hypotheses above, it is further proposed that
H3: Investor sentiment plays a mediating role in the impact of media coverage on the stock market.
4. Study Design
4.1. Sample selection and data sources
This paper selects Chinese listed companies on the Shanghai Stock Exchange and Shenzhen Stock
Exchange since their inception from 2013.1.1 to 2020.1.1 as data samples, and the data time of this
paper ends at the beginning of 2020 to better circumvent the impact of the outbreak of the epidemic
on the Chinese economic market in 2020. In addition, for the convenience of data processing, the
research period of this paper is quarterly, and all individual stock trading data are unified into
quarterly data. Finally, there are a total of 71512 valid data involved in the regression.
We get the data from the Chinese Research Data Services (CNRDS) Platform for media reports,
stock turnover rates, and individual stock returns used in this article.
The following describes the retention and exclusion of data in this paper.
1. In this paper, the number of media reports is selected as a proxy variable for the most media
attention. For the selection of media attention refers to Zhang, Lijuan[21], the field of "total number
of news about the company appearing in the content" is selected from the quantitative statistics of
newspaper financial news.
2. This paper uses individual stock turnover as a proxy for the sentiment of investors[22].
3. The stock codes disclosed in the individual stock return data are used as the benchmark, and the
stock codes that do not correspond to them in other statistics are excluded. In this paper, missing
values are manually removed when collating individual stock return data, and other missing values
are automatically excluded from the regression after combining the data.
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
121
4. In this paper, only the data with the trade status of "successful" are retained to better reflect the
status of stock trading.
4.2. Variable Definition
4.2.1. Explanatory Variables - Media Attention (News).
With the development of Internet technology, the coverage and influence of media have increased,
and the greater the intensity of media coverage proves the higher the attention. the CNRDS database
covers financial news data of listed companies from more than 400 online media and more than 600
newspaper publications, including twenty mainstream online financial news media and eight
mainstream financial newspapers. we download the processed data from the database to avoid any
omission in manual collation. The daily news is processed into quarterly news data, and the data of
media reports are retained according to the valid stock codes. Considering the number of companies
reported is equal to zero and the normality of the data distribution, the number of media reports is
taken as a logarithm plus one.
News = ln (news+1)
4.2.2. Explanatory variable - individual stock return (Rate).
Since there is a lag in investor sentiment feedback to the media in the stock market, individual stock
returns are selected as a proxy for stock market volatility. In this paper, we exclude data with a trading
status other than "normal trading".
4.2.3. Intermediary Variables - Investor Sentiment (Turnover).
As a proxy variable representing market liquidity, the higher the investor sentiment, the greater the
stock turnover rate, and the lower the investor sentiment, the smaller the stock turnover rate. In this
paper, the individual stock turnover rate is used to represent investor sentiment.
4.2.4. Control variables.
Total company size(Size). The larger the company, the more public information is available, the lower
the information asymmetry between managers and investors, and the more stable source of returns.
And because investors are risk averse, they prefer to choose larger companies with better stability
and profitability when making stock investments.
Size = ln (asset)
Total Market Capitalization Outstanding (Mkt cap). Generally, the larger the company is, the higher
the price of the stock in circulation and the larger the market capitalization in circulation. Therefore,
the total outstanding market capitalization is also one of the indicators to judge the profitability of a
company, and investors prefer to choose stocks with a larger total outstanding market capitalization.
Market capitalization = outstanding share capital * stock market price
Book-to-market ratio B/M(B/M). B/M is equal to the book value divided by the market value. Fama
and French concluded that there is a book-to-market effect in the stock market and one of the stock
selection criteria for BM value investors, is that stocks with smaller B/M are growth stocks and stocks
with larger B/M are value stocks, which investors consider more worthy of investment.
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
122
CPI. The composite indicator of investor sentiment also contains macro factors. The consumer price
index, as a macroeconomic indicator, reflects the changes in the purchasing power and consumption
level of China's residents and is an important reference indicator for making macroeconomic
decisions. The CPI level also influences the changes in the capital market at the macro level.
5. Empirical model construction
5.1. Descriptive Statistics
Table 1: descriptive statistics.
Stats
news
Turn over
RATE
B/M
CPI
Size
Mkt cap
N
73495
85668
83384
85590
85668
85592
85668
Min
0.560
0.0298
-0.0121
0.0160
101.2
19.58
19.67
Mean
0.818
1.718
0.00129
0.461
102.0
22.22
22.20
p50
0.731
1.252
0.000103
0.351
102.1
22.01
22.13
p50
0.731
1.252
0.000103
0.351
102.1
22.01
22.13
Max
1.965
7.506
0.0436
3.081
102.9
27.10
25.57
SD
0.222
1.503
0.00679
0.422
0.432
1.428
1.106
Table 1 presents descriptive statistics for variables other than time and stock code. The mean value
of the number of media reports is 0.818 and the median value is 0.731, indicating that there are
differences in the intensity of media coverage of different companies. The maximum value of the
turnover rate is 7.506, the minimum value is 0.296, and the standard deviation is 1.502, indicating
that investor sentiment is more volatile across time, with a mean value of 1.718, indicating that
investor sentiment is relatively high during the period 2013-2019, market liquidity is high, and retail
investors have more uncertainty when trading, which also indicates the lack of perfection in our stock
market. The maximum value of the stock return is 0.436 and the minimum value is -0.121, indicating
that the return profit and loss is unstable, which is in line with the volatile characteristics of the stock
market. The minimum value of company size (Size) is 19.576 and the maximum value is 27.101, and
the minimum value of the book-to-market ratio (B/M) is 0.160 and the maximum value is 3.081, the
data indicate that the size and value of the listed companies vary widely, but the fixed size of
companies is not easy to change in the short term. In addition, the maximum value of CPI is 101.2,
the minimum value is 101.2, and the mean value is 101.990, which indicates that the level of
consumption of the population has been growing during 2013-2019, which is in line with the well-
developed economic status of China. However, due to the small growth of CPI and the two-way fixed
effects applied in the regression model of this paper, CPI is automatically ignored in the regression,
and to avoid the existence of multicollinearity among variables, the correlation analysis of variables
is conducted in this paper, and none of them has multicollinearity, and the maximum absolute value
of CPI correlation coefficient is 0.248. However, it is still retained in this paper, because the
residential consumer index is still However, it is still retained in this paper because the residential
consumer index is still one of the important reference indicators of stock market changes.
5.2. Construction of the empirical model
Rate = c-news +α1 Mkt cap+α2 CPI+α3 Size+α4 B/M+ e1 (1)
Turnover = a(news) +α1 Mkt cap+α2 CPI+α3 Size+α4 B/M + e2 (2)
Rate=c'(news) + b(Turnover) +α1 Mkt cap+α2 CPI+α3 Size+α4 B/M + e3 (3)
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
123
This paper utilizes a three-step regression method stepwise regression, (1) regression of the
explanatory variable news on the explanatory variable Rate (2) showing the regression of the
explanatory variable news on the intermediate variable turnover, and (3) showing the regression of
the explanatory variable news and the intermediate variable turnover on the explanatory variable Rate.
a,b,c,c' are the respective regression coefficients of the various regression coefficients of the core
variables.
According to Wen, C.F.[23], the derivation process of the mediating effect is as follows.
1. Test coefficient c, if it is significant, should be used to argue for a mediating effect, otherwise, it
should be used to argue for a suppressing effect.
2. The coefficients a and b are tested in turn. If they are significant, the indirect effect ab is significant
and the confidence interval is reported. If at least one of them is not significant, then test ab by the
bootstrap method, and if ab is still not significant, then the indirect effect is not significant, and
vice versa.
3. Test the direct effect c', significant then the direct effect is significant, insignificant then the full
mediation effect holds.
4. Testing ab and c' under the premise that c' is significant, if ab and c' have the same sign, the partial
mediating effect holds and the report accounts for ab/c, if ab and c' have different signs, the
suppressing effect holds and the report accounts for ∣ab/c'∣.
5.3. Regression results
The coefficients in the regressions were subjected to 1%-99% tailoring, and *, **, and *** indicate
significant at 1%, 5%, and 10% significance levels, respectively (two-tailed). Since the coefficients
of media coverage on stock reporting rate in both the first and third regressions are 0.005 with three
decimal places retained, it is not possible to conclude that the partial mediation effect holds, so their
regression coefficients are retained with four decimal places.
Table 2: Stepwise regression of mediating effects and bootstrap test
(1)
(2)
(3)
Rate
Turnover
Rate
News
0.0054***
0.484***
0.0048***
(0.000)
(0.035)
(0.000)
Mkt cap
-0.001***
0.395***
-0.002***
(0.000)
(0.023)
(0.000)
CPI
0.000
0.000
0.000
(...)
(...)
(...)
B/M
0.000
-0.783***
0.001***
(0.000)
(0.058)
(0.000)
Size
-0.001***
-0.389***
-0.000***
(0.000)
(0.028)
(0.000)
Turnover
0.001***
(0.000)
_cons
0.045***
1.564**
0.041***
(0.002)
(0.608)
(0.002)
Bootstrap test confidence interval
0.0001719-
0.0002848
0.0031129-
0.0038249
N
71512.000
73373.000
71512.000
r2
0.298
0.524
0.332
Standard errors in parentheses
* p < 0.1,** p < 0.05,*** p < 0.01
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
124
From the regression results of model (1), the total effect c=0.0054, which is significant at the 1%
level of significance, so the theory is established according to the mediating effect. From the
regression results of model (2), a=0.484, which is significant at 1% level of significance, and from
the regression results of model (3), b=0.001 and c'=0.0048, both significant at 1% level of significance
and the confidence interval does not include 0, so the indirect effect is significant. H1 and H2 are
supported. the direct effect is 0.0048 and indirect effect is 0.00484, and c'<c, so the partial mediating
effect holds, and the mediating effect accounts for 11.11%, which supports H3.
In the regression of media coverage, the regression coefficient of the model (1) is 0.0054, model
(2) is 0.484, and model (3) is 0.0048, indicating that an increase in the number of news reports can
cause an increase in stock returns and that media coverage has the most significant effect on investor
sentiment. The regression coefficients of the model (3) are 0.0048 and 0.001, which decrease after
adding mediating variables, indicating that the transmission mechanism weakens the direct effect of
media coverage and investor sentiment on the stock market. According to the partial mediating effect,
there is a transmission relationship between media coverage, investor sentiment, and stock market
changes, media coverage reduces information asymmetry to some extent, increases investors'
decision advantage, and reduces investor sentiment acting on investment, indicating the strong
influential role of media, the function of self-regulation and external regulation of investors and stock
market response to external factors.
6. Conclusion
This paper discusses media coverage, investor sentiment, and stock returns together and examines
the transmission of the relationship between the three. It also investigates that investor sentiment
plays a partially mediating role in the influential relationship between media coverage and stock
market changes using a mediating effect model. This shows that media reports may cause fluctuations
in investor sentiment and capital markets, so attention should be paid to news content so that investors
are more rational in the investment process and stocks fluctuate within a reasonable range.
References
[1] TETLOCK P C. Giving content to investor sentiment: The role of media in the stock market [J]. The Journal of
finance, 62(3): 1139-68 (2007).
[2] GARCIA D. Sentiment during recessions [J]. The journal of finance, 68(3): 1267-300 (2013).
[3] HONG H, STEIN J C. Disagreement and the stock market [J]. Journal of Economic perspectives, 21(2): 109-28
(2007).
[4] FRANKEL R, LI X. Characteristics of a firm's information environment and the information asymmetry between
insiders and outsiders [J]. Journal of accounting and economics, 37(2): 229-59 (2004).
[5] Xiong Yan, Li Changqing, Wei Zhihua. Media coverage and IPO pricing efficiency:Based on information
asymmetry and behavioral finance perspective [J]. World Economy, 37(05): 135-60 (2014).
[6] KIM S-H, KIM D. Investor sentiment from internet message postings and the predictability of stock returns [J].
Journal of Economic Behavior and Organization, 107 (2014).
[7] HANNA A J, TURNER J D, WALKER C B. News media and investor sentiment during bull and bear markets [J].
The European Journal of Finance, 26(14): 1377-95 (2020).
[8] ENGELBERG J E, PARSONS C A. The causal impact of media in financial markets [J]. The Journal of Finance,
66(1): 67-97 (2011).
[9] Zhong Meiling. The impact of financial uncertainty and investor sentiment on stock market volatility [D]; Zhongnan
University of Economics and Law (2020).
[10] SAYIM M, RAHMAN H. The relationship between individual investor sentiment, stock return and volatility:
Evidence from the Turkish market [J]. International Journal of Emerging Markets, 10(3): 504-20 (2015).
[11] PH H, RISHAD A. An empirical examination of investor sentiment and stock market volatility: evidence from India
[J]. Financial Innovation, 6(1): 1-15 (2020).
[12] BAKER M, WURGLER J. Investor sentiment and the cross-section of stock returns [J]. The journal of Finance,
61(4): 1645-80 (2006).
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
125
[13] BAEK C. Stock prices, dividends, earnings, and investor sentiment [J]. Review of Quantitative Finance and
Accounting, 47: 1043-61 (2016).
[14] Chen Q-A, Lei S-Y. Monetary policy, investor sentiment and Chinese stock market volatility: theory and empirical
evidence [J]. China Management Science, 25(11): 1-11 (2017).
[15] Zhong Meiling. The impact of financial uncertainty and investor sentiment on stock market volatility [D]; Zhongnan
University of Economics and Law (2020).
[16] Yin, H., Wu, X. Y.. The predictive role of investors' high-frequency sentiment on intraday stock returns [J]. China
Industrial Economics, (08): 80-98 (2019).
[17] SCHMELING M. Investor sentiment and stock returns: some international evidence [J]. Journal of empirical
finance, 16(3): 394-408 (2009).
[18] FAMA E F. Efficient capital markets: A review of theory and empirical work [J]. The journal of Finance, 25(2):
383-417 (1970).
[19] AKERLOF G A. The market for "lemons": quality uncertainty and the market mechanism [M]. Uncertainty in
economics. Elsevier.235-51(1978).
[20] MASLYUK-ESCOBEDO S, ROTARU K, DOKUMENTOV A. News sentiment and jumps in energy spot and futures
markets [J]. Pacific-Basin Finance Journal, 45: 186-210 (2017).
[21] Zhang Lijuan. Media Attention, Investor Confidence and Corporate Cost of Equity Capital [D]; Shanxi University
of Finance and Economics (2019).
[22] BAKER M, STEIN J C. Market liquidity as a sentiment indicator [J]. Journal of financial Markets, 7(3): 271-99
(2004).
[23] Wen ZL, Ye BJ. Mediated effects analysis: methodological and model development [J]. Advances in Psychological
Science, 22(05): 731-45 (2014).
Proceedings of the 7th International Conference on Economic Management and Green Development
DOI: 10.54254/2754-1169/28/20231308
126
