Using Twitter trust network for stock market analysis

A recent article in this journal by Tabak (2006) highlighted a potentially serious source of bias that can arise when multiple statistical tests of a damages theory are performed and even one test rejecting the null hypothesis is regarded as supporting the damages theory. Repeated testing will eventually produce a “false discovery,” that is, rejection of the null hypothesis in favor of the alternative hypothesis when the null hypothesis is true, which statisticians refer to as type I error. Consequently, performing multiple tests without adjusting the critical value can be problematical because it can lead to improperly accepting statistical evidence that apparently supports rejection of the null hypothesis as reliable when it is not. Tabak (2006) recommends making the Sidak (1968, 1971) multiple-comparison adjustment to the standard statistical t-test to correct for the false-discovery bias inherent in multiple-comparison testing. In particular, he recommends making this adjustment when performing 10b-5 securities fraud event studies when more than one corrective disclosure date is involved. This article clarifies the circumstances in which a multiple-comparison adjustment is appropriate and explains why the correction is normally not needed in securities fraud event-study testing. More generally, I explain why it is not required when each of several tests is performed and its results are reported separately, as for example, where the objective is simply to test the statistical significance of the abnormal stock return on each day on which a new and distinct curative disclosure occurred. I show that the Sidak multiple-comparison adjustment is nearly as stringent as the classical Bonferroni procedure (Simes, 1986), which can increase the risk of type II error. This article discusses a more powerful alternative to the Sidak adjustment due to Benjamini and Hochberg (1993), which directly corrects for the false-discovery bias in multiple-comparison testing and reduces the risk of type II error. II. Application of Multiple-Comparison Adjustments to Securities Fraud Event Studies Multiple-comparison-false-discovery bias can arise when (a) several statistical tests are performed on subsets of the same larger data set in an effort to

This paper conducts a comparative analysis on the variability of firm-level abnormal returns in Egypt and the United States (US) amidst three different recent global hazardous events (political, economic, and health). Moreover, it determines the impact of firm-level characteristics in reducing firms’ abnormal returns and identifies the most vulnerable industries during crises in both countries. The paper methodology is conducted in two stages on the selected sample for US and Egypt stock markets, DJIA and EGX30 indices, respectively. The first stage is an event study approach to determine the amplitude and direction of the response of the US and Egypt abnormal returns to each of the studied crises. The second stage encompasses cross-sectional regressions to study the impact of firm-level characteristics and vulnerable industries on the highest significant abnormal stock returns for each event in each country. This study has theoretical and practical contributions. To our knowledge, this is the first study to demonstrate the higher abnormal stock returns of the developing stock market, Egypt, compared to the developed stock market, US, in both health and economic crises under all different time intervals except the examined political crisis. Moreover, emerging and developed countries’ businesses should provide attentive management of this study’s significant firm-level characteristics to strengthen firms’ exogenous shock absorbency. Furthermore, this study’s identification of the most vulnerable industries in each market paves the way for developing supportive policies for these industries resulting in healthier and safer financial markets amidst crises. Keywords: Abnormal Returns, Stock Markets, Crises, Egypt, United States

This study develops a structural equation model linking both abnormal stock returns and abnormal trading volume to unexpected changes in accounting ratios that result from the issuance of annual accounting data. The three contributions of this study are as follows: (a) a simultaneous investigation of both the price and the trading activity stock market reactions to the issuance of annual financial statements; (b) the application of a measurement model to depict the changes in a firm's financial characteristics, which result from the issuance of financial statements; and (c) a causal model structure is developed, estimated, and tested. Through overidentification of the model, the hypothesized model structure is evaluated. Instead of investigating the price and trading activity reactions separately, as was done in previous studies, the two reactions are modeled simultaneously and are linked. A measurement model is hypothesized to aggregate the unexpected changes in 13 financial ratios into the unexpected changes in 4 latent financial dimensions: liquidity, leverage, profitability, and activity. The structural model then links the unexpected changes in the financial dimensions to abnormal stock market returns and trading volume. The 13 equations of the measurement model and the 2 structural equations are estimated simultaneously using LISREL, a full information maximum likelihood estimation procedure. The model is estimated, and the hypothesized model configuration recreates 82% of the generalized variance in the observed data. The individual parameter estimates show that the major source of variation in abnormal returns is the unexpected changes in profitability. None of the unexpected changes in the financial dimensions are found to be significantly linked directly to abnormal trading volume. Instead, the link is indirect through the relationship between abnormal returns and abnormal volume.

This study aims to analyze the influence of Internet Financial Reporting (IFR) , Level Disclosure and Timeliness (Timelines) Submission of Financial Information Website On The Stock Market .Sample in this study consisted of 70 companies listed in the Jakarta Islamic Index (JII) from 2008-2012. Data analysis uses multiple regression analysis.The results of hypothesis testing showed that the IFR has a significant influence on the abnormal stock returns, the level of disclosure has negative effect on abnormal stock returns and timeliness has no significant effect on abnormal stock returns.

Stock volatility is influenced by the release, dissemination, and acceptance of information. Rumor clarification is expected to reduce asymmetric information and abnormal stock returns by increasing information transparency. However, investors are irrational, and modern behavioral finance studies attribute non-random stock movements to investors’ cognitive and emotional biases. The verification of rumor authenticity may cause fluctuations in investor sentiment, which increases impulsive investing behaviors and stock movements. Due to the widespread and fast accessibility of social media, many electronic information platforms have been established to clarify rumors. It is critical to understand the effects of digitalized rumor clarification on stock markets. In this study, we extracted 12,663 rumor-clarification pairs from 1,804,520 social media posts. We quantified the language used in these messages via sentiment analysis, along with online firm behaviors, to study the effect of clarifications on stock markets. Our findings are as follows: (1) Digitalized rumor-clarification messages affect the abnormal returns of relevant stocks. (2) This influence can be quantified and measured by the emotion polarity of rumor clarification. (3) Firms’ online clarification behaviors, including information disclosure frequency, response time, and wording, have limited to no influence on abnormal returns.

Research Objective - This research was conducted to test whether the government's announcement about Covid-19 was responded to by the capital market with the differences in abnormal returns and stock trading volume before and after the announcement. Methods - The research sample was 660 public companies, listed on the Indonesia Stock Exchange in 2020. The Wilcoxon sign rank test was used to test the research hypothesis, because the data was not normally distributed. Findings – Abnormal stock returns have been proven to have significant differences before and after the Covid-19 announcement. Even stock trading volume activity was shown the same result. The results show significant differences before and after the Covid-19 announcement. The test results for each sector show different abnormal returns before and after the announcement of Covid-19 in the various industries, consumer goods, and basic & chemical industries, while for trading volume activity there are differences only in the consumer goods sector. Theoretical and Policy Implications – The results can be used for decision making for policy makers to resolve the impact of the pandemic, especially on the sectors most affected by the pandemic. The research can also help potential investors in making investment decisions in the future. Research Novelty – This research was conducted on all companies listed on the Indonesian Stock Exchange in 2020 to see the impact of Covid-19 on the entire industrial sector due to the pandemic.

Objective: to examine the effect of signaling the characteristics of business combinations through material facts on the abnormal returns of companies listed on the Brazilian stock exchange. Methods: the study’s final sample comprises 675 observations from 2010 to 2021, collected from the Refinitiv database and the Securities and Exchange Commission website. To analyze the impact on abnormal returns, calculated using the market model and the asset pricing model, the event study methodology was applied with an 11-day event window. Results: the characteristics of business combinations are significant for the Brazilian stock market, resulting in positive abnormal returns. However, no distinctions were observed regarding the type of combination, payment method, or whether a subsidiary conducted the transaction. Notably, when goodwill is signaled, abnormal returns are positive and higher than those without disclosure. Conclusions: The characteristics of business combinations are relevant to the Brazilian stock market, with the information signaled to investors being more impactful than the type of information disclosed. This research highlights the importance of signaling business combination characteristics for managers and informs external users when speculating on such transactions.

The wireless sensor network is an emerging technology, which is used to sense and monitor the environment. As the nodes are deployed in an open environment, the security is one of the essential factors. The cryptography techniques can ensure confidentiality, integrity and authentication. However, wireless sensor network also needs to deal with inside and outside attackers. To deal with outside attackers, attacks by compromised or malicious nodes, trust management system is suggested by many researchers in the area of wireless sensor network. Trust management system can be implemented in various applications for security management such as secure data aggregation, secure cluster head selection, trusted routing, access control, etc. Many researchers provide different kind of solutions for these secure applications based on trust management. However, to incorporate, all such applications on a single sensor node in the network, it is essential to design and develop a trust management system, which considers various aspects and applications of wireless sensor network. As a result, in this paper, we would like to propose a parameter and trust factor based secure communication framework and design a trust management system for wireless sensor networks. Our main contribution is to identify various parameters and trust factors which influences on trust in wireless sensor network and developing a framework for a trust management system based on various parameters and trust factors. The working of the proposed model is shown by simulation experiments conducted in MATLAB for the application of secure communication, data aggregation and intrusion detection in wireless sensor networks.

PurposeThe purpose of this paper is to examine stocks that are most actively discussed by online posters and see if the messages posted about these stocks have information or if they are just noise.Design/methodology/approachThis study uses messages posted on TheLion.com, which reports a real time list of the ten most actively discussed stocks. The stocks in this list at the daily market close during 2005‐2006 are examined. An event study is performed to estimate the daily abnormal returns on these stocks. Contemporaneous and lead–lag regressions of abnormal returns against message posting activities are performed.FindingsOnline posters prefer thinly traded micro‐cap stocks. On average, there is an abnormal return of 19.4 per cent on a stock the day it is one of the ten most talked about stocks. The number of messages posted about a stock on a given day is not only positively related with the stock's abnormal return on that day but it also positively predicts the next day's abnormal return.Research limitations/implicationsIt may be interesting to examine if the investor sentiment expressed in online messages has predictive power for micro‐cap stocks.Practical implicationsThe results provide evidence to regulators that online talk affects stock prices. They show investors that there are inefficiencies in the stock market. They also suggest that corporate managers, especially of small firms, should monitor the stock message boards.Originality/valueThis study focuses on the micro‐cap stocks favored by online posters and finds that online talk has the power to predict the next‐day returns.

This study examined the impact of deceptive advertising on the abnormal stock returns of firms. Using an event study analysis with 101 cases from the FTC database over the period 1987–2005, the FTC rulings on deceptive advertising were found to have the negative effects on the abnormal stock returns of firms. Among the firm-specific factors examined in this study, the amount of advertising expenditures played a role in alleviating the impact of deceptive advertising on the abnormal stock returns, and the firms charged with consent agreements alone were found to lose less firm value than those charged with additional actions by the FTC. Results also showed that the negative effects on the abnormal stock returns created by the FTC actions did not quickly disappear afterwards. These results imply that marketing managers should exercise caution in designing advertising messages that may or may not intend to violate the FTC rules and regulations on deceptive advertising.

The effect of social media sentiments on stock market returns is well-established. However, the quality of content and expertise of content creators vary on social media platforms, and the stocks vary in characteristics. In this research, we examine the effect of sentiment expressed in free content from a social media platform on stock abnormal returns. We also examine the moderating effect of the market capitalisation of stocks on the strength of this relationship. Using data collected from a well-known equity research platform, we demonstrate that the size of the market cap plays an important role in this relationship. The smaller the market cap, the higher the predicting power of the social media sentiment on stock abnormal returns. Considering different holding periods from 1 month to 1 year, we show that sentiments from social media have a long wear in effect on stock abnormal returns. Our results shed light on the importance of market cap and holding period when studying the effect of social media sentiments on stock market returns.

The aim of this study is to analyze and compare the abnormal earnings announcement returns of «concentrated» and «diversified» U.S. REITs from 2000 to 2017. Using a unique property-level dataset, we analyze the impact of geographic concentration on the relative performances of real estate investments trusts (REITs). More precisely, we focus on financial analysts’ forecast accuracy, market-level uncertainty, REIT-level uncertainty and synchronicity. First, we document the coverage, the accuracy and the bias of financial analysts’ earnings forecasts on «concentrated» and «diversified» REITs. Our results report that the level of accuracy and the level of optimism are statistically different for these two categories. Second, we observe that abnormal stock returns, abnormal trading volume and abnormal volatility may be related to the level of geographic concentration and synchronicity. Our results shed a new light on the potential link between the level of geographic concentration, or home bias at home, and the level of information on stock markets.

A stock split is a corporate action that results in a change in the nominal value per share and increases the number of outstanding shares in accordance with the split factor. This study aims to test and analyze whether there are differences in abnormal stock returns and trading volume activity between before and after the stock split event during the Covid-19 pandemic for companies on the Indonesia Stock Exchange. This study uses a quantitative approach with the event study method and a window period of 15 days. The research sample was based on a purposive sampling technique, namely 10 companies out of 12 population companies that had stock splits on the Indonesia Stock Exchange. The analysis technique uses the mean difference test technique, namely the paired sample t-test for abnormal returns, and the Wilcoxon signed rank test for trading volume activity. The results showed that there were differences in abnormal stock returns and trading volume activity before and after the stock split. The average value of abnormal stock returns and trading volume activity before the stock split was higher than after the stock split. Research has implications for all parties as input and consideration in making decisions amid uncertain market conditions.

Stock market trend forecasting is a valuable and challenging research task for both industry and academia. In order to explore the influence of stock news information on the stock market trend, a textual embedding construction method is proposed to encode multiple textual features, including topic features, sentiment features, and semantic features extracted from stock news textual content. In addition, a deep learning method is designed by using financial data and multiple textual features obtained from multiple news textual embeddings for short-term stock market trend prediction. For evaluation, extensive experiments on real stock market data are conducted. The experimental results illustrate that the proposed method can enhance the performance of predicting stock market trend by obtaining effective information from stock news.

The Leverage Effect on Stock Returns