Search results

Since stock return and volatility matters to investors, this study proposes to incorporate the textual sentiment of annual reports in stock price crash risk prediction.

Specific sentences gathered from management discussions and their subsequent analyses are tokenized and transformed into numeric vectors using textual mining techniques, and then the Naïve Bayes method is applied to score the sentiment, which is used as an input variable for crash risk prediction. The results are compared between a collection of predictive models, including linear regression (LR) and machine learning techniques.

The experimental results find that those predictive models that incorporate textual sentiment significantly outperform the baseline models with only accounting and market variables included. These conclusions hold when crash risk is proxied by either the negative skewness of the return distribution or down-to-up volatility (DUVOL).

It should be noted that the authors' study focuses on examining the predictive power of textual sentiment in crash risk prediction, while other dimensions of textual features such as readability and thematic contents are not considered. More analysis is needed to explore the predictive power of textual features from various dimensions, with the most recent sample data included in future studies.

The authors' study provides implications for the information value of textual data in financial analysis and risk management. It suggests that the soft information contained within annual reports may prove informative in crash risk prediction, and the incorporation of textual sentiment provides an incremental improvement in overall predictive performance.

This study aims to explore the risk spillover effects among different sectors of the Chinese stock market after the outbreak of COVID-19 from both Internet sentiment and price fluctuations.

The authors develop four indicators used for risk contagion analysis, including Internet investors and news sentiments constructed by the FinBERT model, together with realized and jump volatilities yielded by high-frequency data. The authors also apply the time-varying parameter vector autoregressive (TVP-VAR) model-based and the tail-based connectedness framework to investigate the interdependence of tail risk during catastrophic events.

The empirical analysis provides meaningful results related to the COVID-19 pandemic, stock market conditions and tail behavior. The results show that after the outbreak of COVID-19, the connectivity between risk spillovers in China's stock market has grown, indicating the increased instability of the connected system and enhanced connectivity in the tail. The changes in network structure during COVID-19 pandemic are not only reflected by the increased spillover connectivity but also by the closer relationships between some industries. The authors also found that major public events could significantly impact total connectedness. In addition, spillovers and network structures vary with market conditions and tend to exhibit a highly connected network structure during extreme market status.

The results confirm the connectivity between sentiments and volatilities spillovers in China's stock market, especially in the tails. The conclusion further expands the practical application and theoretical framework of behavioral finance and also lays a theoretical basis for investors to focus on the practical application of volatility prediction and risk management across stock sectors.

The aim of this paper is threefold: (1) to develop a new measure of investor sentiment rational (ISR) of developing countries by applying principal component analysis (PCA), (2) to investigate co-movements between the ten developing stock markets, the sentiment investor's, exchange rates and geopolitical risk (GPR) during Russian invasion of Ukraine in 2022, (3) to explore the key factors that might affect exchange market and capital market before and mainly during Russia–Ukraine war period.

The wavelet approach and the multivariate wavelet coherence (MWC) are applied to detect the co-movements on daily data from August 2019 to December 2022. Value-at-risk (VaR) and conditional value-at-risk (CVaR) are used to assess the systemic risks of exchange rate market and stock market return in the developing market.

Results of this study reveal (1) strong interdependence between GPR, investor sentiment rational (ISR), stock market index and exchange rate in short- and long-terms in most countries, as inferred from (WTC) analysis. (2) There is evidence of strong short-term co-movements between ISR and exchange rates, with ISR leading. (3) Multivariate coherency shows strong contributions of ISR and GPR index to stock market index and exchange rate returns. The findings signal the attractiveness of the Vietnamese dong, Malaysian ringgits and Tunisian dinar as a hedge for currency portfolios against GPR. The authors detect a positive connectedness in the short term between all pairs of the variables analyzed in most countries. (4) Both foreign exchange and equity markets are exposed to higher levels of systemic risk in the period of the Russian invasion of Ukraine.

This study provides information that supports investors, regulators and executive managers in developing countries. The impact of sentiment investor with GPR intensified the co-movements of stocks market and exchange market during 2021–2022, which overlaps with period of the Russian invasion of Ukraine.

Although risk culture is a key determinant for an effective risk management, identifying the risk culture of a firm can be challenging due to the abstract concept of culture. This paper proposes a novel approach that uses unsupervised machine learning techniques to identify significant features needed to assess and differentiate between different forms of risk culture.

To convert the unstructured text in our sample of banks' 10K reports into structured data, a two-dimensional dictionary for text mining is built to capture risk culture characteristics and the bank's attitude towards the risk culture characteristics. A principal component analysis (PCA) reduction technique is applied to extract the significant features that define risk culture, before using a K-means unsupervised learning to cluster the reports into distinct risk culture groups.

The PCA identifies uncertainty, litigious and constraining sentiments among risk culture features to be significant in defining the risk culture of banks. Cluster analysis on the PCA factors proposes three distinct risk culture clusters: good, fair and poor. Consistent with regulatory expectations, a good or fair risk culture in banks is characterized by high profitability ratios, bank stability, lower default risk and good governance.

The relationship between culture and risk management can be difficult to study given that it is hard to measure culture from traditional data sources that are messy and diverse. This study offers a better understanding of risk culture using an unsupervised machine learning approach.

The purpose of this study is to assess how the information disclosed in prospectuses impacted the initial public offering (IPO) underpricing at a time of high government interference amid the ongoing pandemic.

The design of this study has several tracks, namely, a macro-level track, which is represented by the government measures to halt the pandemic; a micro-level track, which is followed by textual analysis of IPO prospectuses; and, finally, a machine learning track, in which the authors use state-of-the-art tools to improve their linear regression model.

The authors found that strict government anti-COVID-19 measures indeed contribute to the reduction of the IPO underpricing. Interestingly, the mere fact of such measures taking place is enough to take effect on financial markets, regardless of the resulting efficiency of such measures. At the micro-level, the authors show that prospectus sentiments and their significance differ across prospectus sections. Using linear regression and machine learning models, the authors find robust evidence that such sections as “Risk factors”, “Prospectus summary”, “Financial Information” and “Business” play a crucial role in explaining the underpricing. Their effect is different, namely, it turns out that the more negative “Risk factors” and “Financial Information” sentiment, the higher the resulting underpricing. Conversely, the more positive “Prospectus summary” and “Business” sentiments appear, the lower the resulting underpricing is. In addition, we used machine learning methods. Consisting of more than 580 IPO prospectuses, the study sample required modern and powerful machine learning tools like Isolation Forest for pre-processing or Random Forest Regressor and Light Gradient Boosting Model for modelling purposes, which enabled the authors to gain better results compared to the classic linear regression model.

At the micro level, this study is not confined to 2020, but also embraces 2021, the year of the record number of IPOs held. Moreover, in this paper, these were prospectuses that served as a source of management sentiment. In addition, the authors used a tailor-made government stringency index. At the micro level, basing the study on behavioural finance hypotheses, the authors conducted both separate and holistic analysis of prospectuses to assess investors’ reaction to different aspects of IPO companies as well as to the characteristics of the IPOs themselves. Lastly, the authors introduced a few innovations to the research methodology. Textual analysis was conducted on a corpus of prospectuses included in a study sample. However, the authors did not use pre-trained dictionaries, but instead opted for FLAIR, a modern open-source framework for natural language processing.

This study quantified companies' views on the COVID-19 pandemic with sentiment analysis of US public companies' disclosures. The study aims to provide timely insights to shareholders, investors and consumers by exploring sentiment trends and changes in the industry and the relationship with stock price indices.

From more than 50,000 Form 10-K and Form 10-Q published between 2020 and 2021, over one million texts related to the COVID-19 pandemic were extracted. Applying the FinBERT fine-tuned for this study, the texts were classified into positive, negative and neutral sentiments. The correlations between sentiment trends, differences in sentiment distribution by industry and stock price indices were investigated by statistically testing the changes and distribution of quantified sentiments.

First, there were quantitative changes in texts related to the COVID-19 pandemic in the US companies' disclosures. In addition, the changes in the trend of positive and negative sentiments were found. Second, industry patterns of positive and negative sentiment changes were similar, but no similarities were found in neutral sentiments. Third, in analyzing the relationship between the representative US stock indices and the sentiment trends, the results indicated a positive relationship with positive sentiments and a negative relationship with negative sentiments.

Performing sentiment analysis on formal documents like Securities and Exchange Commission (SEC) filings, this study was differentiated from previous studies by revealing the quantitative changes of sentiment implied in the documents and the trend over time. Moreover, an appropriate data preprocessing procedure and analysis method were presented for the time-series analysis of the SEC filings.

This paper aims to examine the dynamic conditional correlation (DCC) and hedging ratios between Dow Jones markets and oil, gold and bitcoin. Using daily data, including the COVID-19 pandemic and the Russia–Ukraine war. We employ the DCC-generalized autoregressive conditional heteroskedasticity (GARCH) and asymmetric DCC (ADCC)-GARCH models.

DCC-GARCH and ADCC-GARCH models.

The most of DCCs among market pairs are positive during COVID-19 period, implying the existence of volatility spillovers (Contagion-effects). This implies the lack of additional economic gains of diversification. So, COVID-19 represents a systematic risk that resists diversification. However, during the Russia–Ukraine war the DCCs are negative for most pairs that include Oil and Gold, implying investors may benefit from portfolio-diversification. Our hedging analysis carries significant implications for investors seeking higher returns while hedging their Dow Jones portfolios: keeping their portfolios unhedged is better than hedging them. This is because Islamic stocks have the ability to mitigate risks.

Our paper may make a valuable contribution to the existing literature by examining the hedging of financial assets, including both conventional and Islamic assets, during periods of stability and crisis, such as the COVID-19 pandemic and the Russia–Ukraine war.

This paper aims to examine whether managerial sentiment, extracted from annual reports, is associated with corporate risk-taking in the context of Chinese companies. This study expands the vein of literature on overconfidence theory.

By leveraging textual analysis on Chinese listed companies’ annual reports, the authors construct firm-level managerial sentiment during 2007 and 2021 to examine how managerial sentiment influences corporate risk-taking after control for firm characteristics. Corporate risk-taking is denoted by corporate investment engagements: capital expenditures and net fixed asset investment.

Results show that incentives for corporate risk-taking are likely to increase with the positive managerial sentiment and decrease with the negative sentiment in companies’ annual reports. Positive managerial sentiment is associated with over-/under-investment and low/high investment efficiency. Further additional tests show that the managerial sentiment effect only holds during low economic uncertain years and samples of private-owned firms. Furthermore, the robust tests indicate that there is no endogenous issue between managerial sentiment and corporate risk-taking.

Annual report textual-based managerial sentiment may not perfectly reflect managers’ lower frequency sentiment (e.g. weekly, monthly and quarterly sentiment). Future studies could attempt to capture managers’ on-time sentiment by using media sources and corporate disclosures.

To the best of the authors’ knowledge, this paper is the first research to provide insights into supervising managers’ corporate decisions by observing their textual information usage in corporate disclosure. Moreover, the approach of measuring managerial sentiment might be a solution to monitoring managerial class.

This paper contributes to the literature on accounting and finance studies, adding another piece of empirical evidence on content analysis by examining a unique language and institutional context (i.e. China). Besides, the paper notes that in line with the English version disclosure, based on Chinese semantic words, managerial sentiment in the Chinese-speaking world has magnitude on corporate decisions. The research provides insights into supervising managers’ corporate decisions by observing their textual information usage in corporate disclosure. Moreover, the approach to measuring managerial sentiment may be a practical solution to monitoring managerial class.

Social media data from financial websites contain information related to enterprise credit risk. Mining valuable new features in social media data helps to improve prediction performance. This paper proposes a credit risk prediction framework that integrates social media information to improve listed enterprise credit risk prediction in the supply chain.

The prediction framework includes four stages. First, social media information is obtained through web crawler technology. Second, text sentiment in social media information is mined through natural language processing. Third, text sentiment features are constructed. Finally, the new features are integrated with traditional features as input for models for credit risk prediction. This paper takes Chinese pharmaceutical enterprises as an example to test the prediction framework and obtain relevant management enlightenment.

The prediction framework can improve enterprise credit risk prediction performance. The prediction performance of text sentiment features in social media data is better than that of most traditional features. The time-weighted text sentiment feature has the best prediction performance in mining social media information.

The prediction framework is helpful for the credit decision-making of credit departments and the policy regulation of regulatory departments and is conducive to the sustainable development of enterprises.

The prediction framework can effectively mine social media information and obtain an excellent prediction effect of listed enterprise credit risk in the supply chain.

The purpose of this study is to examine the relationship between explanatory language, audit fees and audit hours to demonstrate that auditors use explanatory language in audit reports to explain perceived audit risk.

The authors construct the sentiment value, a novel audit risk proxy derived from audit reports, using big data analysis. The relationship between sentiment value and explanatory language is then investigated. The authors present the validity of their new metric by examining the relationship between sentiment value and accounting quality, taking audit fees and hours into account.

The authors first find that reporting explanatory language is positively related to audit fees. More importantly, the authors provide an evidence that explanatory language in audit reports is indicative of increased audit risk as it is negatively correlated with sentiment value. As a positive (negative) sentimental value means that the audit risk is low (high), the results indicate that auditors describe explanatory language in a negative manner to convey the inherent audit risk and receive higher audit fees from the risky clients. Furthermore, the relationship is strengthened when the explanatory language is more severe, such as reporting the multiple numbers of explanatory language or going-concern opinion. Finally, the sentiment value is correlated with accounting quality, as measured by the absolute value of discretionary accruals.

Contrary to previous research, the authors’ findings suggest that auditors disclose audit risks of client firms by including explanatory language in audit reports. In addition, the authors demonstrate that their new metric effectively identifies the audit risk outlined qualitatively in audit report. To the best of the authors’ knowledge, this is the first study that establishes a connection between sentiment analysis and audit-related textual data.