scholarly journals Cross-Sectional Absolute Deviation Approach for Testing the Herd Behavior Theory: The Case of the ASE Index

Author(s):  
Imad Zeyad Ramadan
2018 ◽  
Vol 10 (3(J)) ◽  
pp. 203-219 ◽  
Author(s):  
Kalugala Vidanalage Aruna Shantha

This paper examines the herding phenomenon in the context of a frontier stock market, the Colombo Stock Exchange of Sri Lanka, employing the cross - sectional absolute deviation methodology to daily frequencies of data for the period from April, 2000 to September, 2016. The results show significant changes in magnitude and pattern of herding over different episodes of the market. The herd behavior is strongly presence irrespective of the direction of the market movement in the 2000 - 2008 period, during which investments in the stock market is affected by the country’s political instability resulting from the civil war. The evidence also shows herd behavior during the period of market bubble whereas negative herding in the market crash period. However, it becomes less likely to occur during the period after the market crash. The lower tendency to herd during the post- market crash period supports the Adaptive Market Hypothesis, implying that investors are likely to realize the irrationality of herding and learn to be more rational as a consequence of significant losses experienced during the period of the market crash. Accordingly, these findings suggest that period- specific characteristics of the market and the associated psychological effects to investors such as overconfidence and panics would cause changes to their beliefs and behavior, the experiences of which would subsequently produce learning effect to minimize their irrationality in decision making.


2017 ◽  
Vol 16 (4) ◽  
pp. 497-515 ◽  
Author(s):  
Houda Litimi

Purpose This paper aims to investigate the herding behavior in the French stock market and its effect on the idiosyncratic conditional volatility at a sectoral level. Design/methodology/approach This sample covers all the listed companies in the French stock market, classified by sector, over four major crisis periods. The author modifies the cross-sectional absolute deviation (CSAD) model to include trading volume and investors sentiment as herding triggers. Furthermore, the author uses a modified GARCH model to investigate the effect of herding on conditional volatility. Findings Herding is present in the French market during crises, and it is present in only some sectors during the entire period. The main trigger for investors to embark into a collective herding movement differs from one sector to another. Furthermore, herding behavior has an inhibiting effect on market conditional volatility. Originality/value The author modifies the CSAD model to investigate the presence of herding in the French stock market at a sectoral level during turmoil periods. Furthermore, the particularly designed GARCH model provides new insights on the effect of herding and volume turnover on the conditional volatility.


2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Sunaina Kanojia ◽  
Deepti Singh ◽  
Ashutosh Goswami

PurposeHerd behavior has been studied herein and tested based on primary respondents from Indian markets.Design/methodology/approachThe paper expounds the empirical evidence by applying the cross-sectional absolute deviation method and reporting on herd behavior among decision-makers who are engaged in trading in the Indian stock market. Further, the study attempts to analyze the market-wide herding in the Indian stock market using 2230 daily, 470 weekly and 108 monthly observations of Nifty 50 stock returns for a period of nine years from April 1, 2009 to March 31, 2018 during the normal market conditions, extreme market conditions and in both increasing and decreasing market conditions.FindingsIn a span of a decade witnessing different market cycles, the authors’ results exhibit that there is no evidence of herding in any market condition in Indian stock market primarily due to the dominance of institutional investors and secondly because of low market participation by individual investors.Originality/valueThe results reveal that there is no impact of herd behavior on the stock returns in the Indian equity market during the normal market conditions. It highlights that the participation of individuals who are more prone to herding is more evident for short-run investments, contrary to long-term holdings.


2021 ◽  
Vol 6 (3) ◽  
pp. 17-25
Author(s):  
Muhammad Tayyab Ul Hassan ◽  
Syed Hassan Jamil

This study investigates the influence of herd behavior on the Pakistan stock exchange indexes KSE-100 and KSE-30 during bullish and bearish markets. Using the daily market return from 2007 to 2020. We implement the method of main herding measures, Cross-sectional absolute deviation, and Cross-sectional standard deviation, to explore the influence of herd behavior in the emerging market of Pakistan. The results indicate the presence of market-wide herd behavior: (a) along with the different direction of market positive and negative return, (b) when trading volume high, (c) when stock market highly volatile, and (d) during and the post-financial crisis. Moreover, Investors don’t herd when low trading volume and low volatility. Our study fills the gap in the literature and contributes to academic relevance by exploring the influence of herd behavior among both bull and bear periods in markets of Pakistan, it also examines the possible asymmetric effects of herding related to the market with high-low trading volume and market volatility.


2018 ◽  
Vol 44 (7) ◽  
pp. 919-934 ◽  
Author(s):  
Chun-Da Chen ◽  
Riza Demirer

Purpose The purpose of this paper is to show that the level of herding in an industry can be the basis for a profitable investment strategy. Design/methodology/approach The authors apply three different herding measures in the paper, including cross-sectional standard deviation, cross-sectional absolute deviation and non-linear model – state–space model. Findings The authors find that industries that experience a high level of herding yield higher subsequent returns regardless of their past performance. Consequently, the authors show that a herding-based investment strategy generates significant profits, even after adjusting for risk. The findings also show that the herding effect when combined with past performance as part of a conditional investment strategy yields significant profits regardless of the formation and holding periods. The findings suggest that the level of herding could serve as a systematic driver of returns and could be exploited for profitable investment strategies. Originality/value To the best of authors’ knowledge, this is the first study in the literature to show that herding by itself can serve as a determinant of returns regardless of past performance.


2020 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Mouna Youssef ◽  
Khaled Mokni

PurposeThis study aims to test the presence of herding behavior in commodity markets, including energy, metals and agriculture. Additionally, the authors investigate the possible asymmetric effect of oil price changes on the herding behavior in these markets.Design/methodology/approachThe authors examine herding based on the cross-sectional absolute deviation (CSAD) model in a static and time-varying perspective.FindingsBy using daily data over the period 2003–2017, the authors’ findings firstly support the dynamic nature of investor behavior in commodity markets, which oscillates between antiherding during the normal period and herding during and after the global financial crisis of 2008. Furthermore, results highlight that the asymmetric impact of oil shocks on herding differs across commodity sectors and periods. Additionally, herding seems to be more pronounced when the oil market declines, which may be due to the pessimistic investors' sentiments.Practical implicationsThis study provides insight into what factors influence herd behavior in commodity markets. The understanding of factors driving herding aids investors to avoid the impact of this behavior and its consequencesOriginality/valueTo the authors’ knowledge, this study is the first to examine whether the level of herding depends on the oil price fluctuations, as well as the asymmetric effect of the oil price on herding behavior in commodity markets.


Author(s):  
Made Dewi Ayu Untari

The purposes of this study are to obtain emperical evidence about the influence of followers investor’s behaviour to the stock volatility and analyze the difference offollowers investor’s betweenindustry sectors producing raw materials,manufacture industry and service industry in the Indonesia Stock Exchange (BEI), during the market crash happened in Indonesia. The population number are 507 companies, while the total sample of 247 companies. Sampling technique used purposive sampling. The analysis technique used was a cross-sectional absolute Deviation (CSADand test One Way ANOVA with Post Hoc Test and Least Significant Difference (LSD. Data shows that the behavior of follower investors has positive effect on the volatility of the current stock market crash occurs. Meanwhile, there was no difference in behavior between the follower investor industrial sectors producing raw materials, the manufacturing sector and the service sector when  the market crash.


Author(s):  
Mbarga M. J. Arsene ◽  
Hippolyte T. Mouafo ◽  
I. V. Podoprigora ◽  
L. A. Smolyakova ◽  
N. V. Yashina ◽  
...  

Background: After the worldwide spread of the coronavirus pandemic, several experts predicted a health catastrophe in Africa. However, the expected earthquake ultimately did not occur and the statistics of the number COVID-19 cases and deaths for other continents (Europe, America, Asia) were far higher than those of Africa. This study focused on Central Africa tried to explain this low incidence of COVID-19. Methodology: A cross-sectional time series method was adopted and the data of COVID-19 cases and deaths for Angola, Cameroon, Chad, Central African Republic, Congo Brazzaville, Democratic Republic of Congo, Gabon, Equatorial Guinea and Sao Tome and Principe between March and November 2020 were extracted from the World Health Organization COVID-19 database. The evolution of COVID-19 cases and deaths for each country were plotted and the accuracy measures such as Mean Absolute Percentage Error, Median Absolute Deviation and Mean Squared Displacement were calculated. Association between the countries and the prevalence of cases, deaths and recovered was visualized through principal component analysis. Results: The results showed that the highest number of cases was observed in Cameroon (21,793) while Sao Tome and Principe scored the smallest one (962). However, based on the total population, the prevalence of COVID-19 cases was high in Sao Tome and Principe (0.436%) and Gabon (0.400%). The highest death percentages (≥2%) were observed in Chad (6.742%), RDC (2.708%) and Angola (2.592%) while the highest recovered percentages were in Gabon (99.10%), Equatorial Guinea (97.62%) and Cameroon (97.02%). Development of traditional medicines and modification of food behavior including consumption of plant extracts appear as the reasons for the highest recovered rates. The accuracy measurements showed that the trend curves were not correlated with the actual evolution of the pandemic, but the Spearman correlation test revealed that except Equatorial Guinea (r=0.042, p=0.817), the evolution of COVID-19 cases and deaths were strongly correlated. Conclusion: The overall prevalence and incidence of COVID-19 is low in the countries of the Central Africa sub-region despite the problems facing the health systems of these countries.


2019 ◽  
Vol 11 (1(J)) ◽  
pp. 181-190
Author(s):  
Mustapha El Hami ◽  
Ahmed Hefnaoui

Frontier markets, particularly the Moroccan financial market, are characterized by a narrowness of market, inability to absorb erratic price fluctuations and the low liquidity of securities that encourage investors to herd and imitate those who have all the information about the market. A quantitative research approach was used to analyze the existence of herding n Moroccan stock market. The daily data used in this study concerns the period from 04/01/2010 to 29/12/2017 and contains the daily returns of the MASI and a total of 43 traded stocks. Statistical and econometric methods such as multidimensional scaling and Cross-sectional absolute deviation were used. Subsequently, after the regression models were examined, findings indicated that the first stocks with the highest similarity to the index return are BMCE, BCP, IAM, ATW and CMSR, and the first stocks with the highest dissimilarity are PAP, IBC and SNP, This will have to allow investors to choose profitable alternatives and avoid those that present a possible risk. The results did also show the existence of herding in the Moroccan stock market both upward and downward. This finding was supported by the clear existence of a non-linearity between market performance and CSAD measurement, which confirms the prediction of a non-linear inversion relationship between CSAD and 𝑅𝑚. This could be due to the low level of transparency that prevails in frontier stock exchanges and reduces the quality of their information environment, which leads investors not to react rationally and to draw information from the transactions of their peers.


Sign in / Sign up

Export Citation Format

Share Document