Journal articles on the topic 'Loughran McDonald Sentiment Classifier'

This paper studies how sentiment affect Bitcoin pricing by examining, at an hourly frequency, the linkage between sentiment of finance-related Twitter messages and return as well as the volatility of Bitcoin as a financial asset. On the one hand, there was calculated the return from minute-level Bitcoin exchange quotes and use of both rolling variance and high-minus-low price to proxy for Bitcoin volatility per each trading hour. On the other hand, the mood signals from tweets were extracted based on a list of positive, negative, and uncertain words according to the Loughran-McDonald finance-specific dictionary. These signals were translated by categorizing each tweet into one of three sentiments, namely, bullish, bearish, and null. Then the total number of tweets were adopted in each category over one hour and their differences as potential Bitcoin price predictors. The empirical results indicate that after controlling a list of lagged returns and volatilities, stronger bullish sentiment significantly foreshadows higher Bitcoin return and volatility over the time range of 24 hours. While bearish and neutral financial Twitter sentiments have no such consistent performance, the difference between bullish and bearish ratings can improve prediction consistency. Overall, this research results add to the growing Bitcoin literature by demonstrating that the Bitcoin pricing mechanism can be partially revealed by the momentum on sentiment in social media networks, justifying a sentimental appetite for cryptocurrency investment.

PurposeThis study examines whether language disclosed in the Management Discussion and Analysis (MD&A) of US Real Estate Investment Trusts (REITs) provides signals regarding future firm performance and thus generates a market response.Design/methodology/approachThis research conducts textual analysis on a sample of approximately 6,500 MD&As of US REITs filed by the SEC between 2003 and 2018. Specifically, the Loughran and Mcdonald (2011) financial dictionary, and a custom dictionary for the real estate industry created by Ruscheinsky et al. (2018), are employed to determine the inherent sentiment, that is, the level of pessimistic or optimistic language for each filing. Thereafter, a panel fixed-effects regression enables investigating the relationship between sentiment and future firm performance, as well as the markets’ reaction.FindingsThe empirical results suggest that higher levels of pessimistic (optimistic) language in the MD&A predict lower (higher) future firm performance. Hereby, the use of a domain-specific real estate dictionary, namely that developed by Ruscheinsky et al. (2018) leads to superior results. Corresponding to the notion that the human psyche is affected more strongly by negative than positive news (Rozin and Royzman, 2001), the market responds solely to pessimistic language in the MD&A.Practical implicationsThe results suggest that the market can benefit from textual analysis, as investigating the language in the MD&A reduces information asymmetries between US REIT managers and investors.Originality/valueThis is the first study to analyze exclusively US REITs, whether language in the MD&A is predictive of future firm performance and whether the market responds to textual sentiment.

Recently, the investment information is spreading in near real-time due to the global integration of the international financial market. At this point, it is essential to resolve the information asymmetry in the capital market to vitalize domestic and foreign investment. Analysts improve market efficiency, by performing information intermediary role between the capital market and investors. However, the herding behavior among analysts in the market can weaken the information quality and lead to inefficiency in the financial market. In this study, the relationship between analyst herding behavior and the text of the report is analyzed. In addition, the effect of the text of the report body on the information power according to the analyst's herding behavior was verified. To this end, the positive and negative emotions shown in the body of the analyst report were extracted as words of up to 2-gram units and merged with the list of positive and negative words of Loughran and McDonald(2011), a sentiment dictionary widely used in the financial field. As a result of the analysis, when an analyst presents a 'bold' opinion that deviates from the consensus, the sentiment of the report body is more extreme than that of an ‘herd’ opinion. In addition, it was confirmed that positive emotion weakens the report information power according to the analyst's herding behavior. This study is expected to alleviate information asymmetry between investors and the financial market. Furthermore, the text mining methodology used in this study can be readily applied to any other kind of text, including News, Social Network Services, or IR reports.

Objective: identification of the relationship between the news coverage of global diseases and the dynamics of the return on shares of the pharmaceutical sector for Russia and the United States.Material and methods. The empirical base of the study includes more than 700 thousand tweets on Ebola and COVID-19 in Russian and English, news of the RBC news agency. The sentiment of the text was assessed on the basis of five English and four Russian-language dictionaries, the influence of fundamental and textual variables on the profitability of pharmaceutical companies' shares was carried out using the ARMAX-GARCH econometric model.Results. It has been proven that the dynamics of the stock index of pharmaceutical companies is explained by fundamental (economic) and sentimental factors. News of any epidemics negatively affects the pharmaceutical sector in the US and Russia, that is, there are no industries that benefit from this situation. Pandemic news affects US pharmaceutical companies more than Russian companies. The effect of news influence depends on the level of spread of the disease. News influences not only at the moment of their publication, but also after: there is a "delayed effect". Ebola news affects the American pharmaceutical market for 2 weeks, and the dynamics of the increase in influence can be traced. News on the COVID pandemic amplifies its impact during 1 week for the Russian pharmaceutical market and for 2 weeks for the US pharmaceutical companies. As for news sources, the elastic network has identified more significant variables based on publications from RBC; therefore, Internet publications generate more publicity, shaping a more significant overall sentiment in the markets.Conclusion. The models developed in the framework of the study and the economic conclusions obtained have not only theoretical, but also practical significance, and can also be used for further research in this area. It is possible to give recommendations on the practical use of dictionaries to assess the sentiment of the text. In our study, the elastic network method chose the Loughran–McDonald dictionary for evaluating economic texts in English and the EcSentiThemeLex dictionary (designed in R and Python programming environments). Avenues for further investigation may include analysis of other sources of information about the pandemic.

We compare the ability of dictionary-based and machine-learning methods to capture disclosure sentiment at 10-K filing and conference-call dates. Like Loughran and McDonald [Loughran T, McDonald B (2011) When is a liability not a liability? Textual analysis, dictionaries, and 10-Ks. J. Finance 66(1):35–65.], we use returns to assess sentiment. We find that measures based on machine learning offer a significant improvement in explanatory power over dictionary-based measures. Specifically, machine-learning measures explain returns at 10-K filing dates, whereas measures based on the Loughran and McDonald dictionary only explain returns at 10-K filing dates during the time period of their study. Moreover, at conference-call dates, machine-learning methods offer an improvement over the Loughran and McDonald dictionary method of a greater magnitude than the improvement of the Loughran and McDonald dictionary over the Harvard Psychosociological Dictionary. We further find that the random-forest-regression-tree method better captures disclosure sentiment than alternative algorithms, simplifying the application of the machine-learning approach. Overall, our results suggest that machine-learning methods offer an easily implementable, more powerful, and reliable measure of disclosure sentiment than dictionary-based methods. This paper was accepted by Brian Bushee, management science.

Abstract Growing AI readership (proxied for by machine downloads and ownership by AI-equipped investors) motivates firms to prepare filings friendlier to machine processing and to mitigate linguistic tones that are unfavorably perceived by algorithms. Loughran and McDonald (2011) and BERT available since 2018 serve as event studies supporting attribution of the decrease in the measured negative sentiment to increased machine readership. This relationship is stronger among firms with higher benefits to (e.g., external financing needs) or lower cost (e.g., litigation risk) of sentiment management. This is the first study exploring the feedback effect on corporate disclosure in response to technology.

ABSTRACT This study examines the degree to which Loughran and McDonald (LM) word lists are informative at the item level of SEC filings, such as risk factors (RF) and management’s discussion and analysis (MDA) disclosures in 10-X reports. In this context, we explore if sentiment types are informative when associated with other material events, namely cybersecurity breaches. Our results support the assertion that sentiment types, beyond positive and negative, are informative at the individual disclosure item level, as tested in the RF and MDA sections. We also find that investors respond to different types of sentiment between RF and MDA. We find an economically significant estimated average economic impact of $469 million/firm. We further contribute to the literature by applying novel statistical methods that advance empirical accounting literature. Data Availability: Data are available from the public sources cited in the text.

The media performs an essential role in providing information regarding initial public offerings (IPOs) to investors. Often, public information about IPO companies is scarce and investors find it difficult to comprehend complex documents, making media a cost-effective and authentic source of information. This study explores how media sentiment affects the underpricing and initial aftermarket performance of IPOs using 3,124 media articles published from April 2019 to March 2022. In doing so, the authors employed textual sentiment analysis using the Loughran and McDonald dictionary to deduce the media net sentiment score. Subsequently, robust regression is conducted by incorporating the media sentiment, the volume of media articles, and other control variables. The result mentions that the way media present information has the potential to influence the decision of retail investors. The net sentiment score shows that the positive information in the media article makes investors optimistic about the prospects of the business. It boosts their confidence and creates more demand for shares, thereby increasing the IPO’s first-day market-adjusted abnormal return. Further, it has been discovered that media sentiment significantly influences initial aftermarket performance primarily due to the limited trading history available for the IPO company during the initial stage.

PurposeThe goal of the study is to examine the effects of management discussion and analysis (MD&A) sentiment in public companies' annual reports on corporate investment incentives in developing economies.Design/methodology/approachThe authors use sentiment analysis of MD&A texts based on Loughran and McDonald (2011) and combination of panel data regression, logit model and random forest. The text data consists of 3,511 annual reports of Chinese listed companies for the period from 2010 to 2019.FindingsThis paper provides empirical evidence of signaling theory that sentiment of annual reports and MD&A influences corporate decisions on both M&A and internal investments. The authors found that comparing to annual reports MD&A sentiment has more stable and significant explanatory and predictive power.Practical implicationsThis paper confirms the importance of MD&A sentiment for corporate investment decision taking and provides practical techniques for analysts and researchers to study corporate investment incentives from the point of view of signaling theory.Originality/valueThe study aims to expand the domains of signaling theory and corporate investment valuation by including a broader range of data on companies' M&A and internal investments in developing economies. To explore the impact of MD&A sentiment on corporate investment, a state-of-the-art set of text mining and machine learning techniques is used. The authors' results confirm that MD&A has signaling effect and can get a positive market response. Furthermore, this study enhances the empirical evidence of overconfidence theory, i.e. optimistic management whose MD&A tend to positive overestimates the management's investments decision and also underestimate the potential risk to the firm.

Purpose The purpose of this study is to explore audit procedure disclosures related to key audit risks, during the prior year and the initial year of the COVID-19 outbreak, by reporting on matters published in over 3,000 Australian statutory audit reports during 2019 and 2020. Design/methodology/approach This study partially uses latent semantic analysis methods to apply textual and readability analyses to external audit reports in Australia. The authors measure the tone of the audit reports using the Loughran and McDonald (2011) approach. Findings The authors find that 3% of audit procedures undertaken during 2020 were designed to address audit risks associated with the COVID-19 pandemic. As a percentage of total audit procedures undertaken during 2020, the authors find that smaller practitioners reported much less audit procedures related to COVID-19 audit risks than most larger audit firms. Finally, the textual analysis further found differences in the sentiment or tone of words used by different auditors in 2020, but differences in sentiment or tone were not found when 2020 was compared to the prior year 2019. Originality/value This study provides early evidence on whether auditors designed audit procedures to deal specifically with audit risks that arose due to the COVID-19 pandemic and on the extent and nature of those audit procedures. The study will help policymakers to better understand whether Key Audit Matters provided informational value to investors during a time of global crisis.

Purpose This paper aims to examine if Malaysian public listed companies have expressed any specific sentiment(s) when publishing their financial performance during the COVID-19 pandemic. Design/methodology/approach The disclosed sentiments contained in the management discussion and analysis section of the companies’ annual reports were extracted by means of computer-automated textual analysis through the linguistic inquiry and word counts and the Loughran–McDonald Financial Sentiment Dictionary. Next, a correlation analysis was conducted. Finally, a qualitative content analysis (QCA) was conducted to confirm these sentiments. Findings The analysis shows that companies adopted various tones of sentiments when communicating with their stakeholders. Most companies used negative sentiments to voice their concerns about how the COVID-19 pandemic has impacted upon their business operations. Only a few companies reflected positive sentiments, whilst those that experienced operating losses also expressed uncertainty. Research limitations/implications This study may assist either the regulators or accounting bodies to introduce a reporting framework that public companies can adopt during natural hazards. It also provides useful insights to (potential) investors to enable them to better understand the business landscape. For future research, the same study could be conducted on more countries so that their experiences can be used to better understand the business phenomenon from a global perspective. Originality/value This study is one of few studies to adopt automated textual analysis and QCA to examine the exhibited sentiments when public companies reported their financial performance during the COVID-19 pandemic.

Purpose Drawing on signalling and impression management theories, this study aims to examine a bidirectional association between managerial tone in earnings conference calls and financial performance. Design/methodology/approach The sample includes non-financial firms listed in the FTSE 350 index during the period 2010–2015. Managerial tone was measured using positive and negative keywords based on the Loughran-McDonald Sentiment Word Lists, while return on assets was used as a proxy for firms’ financial performance. Findings The findings indicate that current financial performance positively affects the managerial tone in earnings conference calls. Likewise, the results also show that there is a positive relationship between managerial tone in earnings conference calls and firms’ future financial performance. Practical implications The results have important implications for top management to use more virtual communication media (i.e. earnings conference calls) to continue managing their relationships with financial stakeholders and helping them better understand financial performance, especially in countries where holding such calls is not yet part of firms’ policy. Originality/value To the best of the authors’ knowledge, this is one of the first studies that explore the relationship between managerial tone in earnings conference calls and financial performance. Overall, this study contributes to managerial tone literature and holds significant theoretical and practical implications.

PurposeThis study aims to analyse Moody’s use of sustainable finance keywords in over 24,000 rating action reports from 2012 to 2024. Prior to Moody’s 2019 acquisition of Vigeo Eiris (VE) – a specialist sustainable research firm – sustainable finance keywords were more closely associated with downgrades and conveyed a negative tone. Post-acquisition, the proportion of sustainable finance terminology increased, while its influence on rating outcomes diminished and conveyed a positive tone. These findings reflect broader systemic shifts driven by acquisitions, market forces and regulatory pressures, suggesting a normalization of sustainable finance factors in Moody’s credit evaluations and marking their shift from exceptional to routine considerations.Design/methodology/approachThe study analyses over 24,000 Moody’s rating action reports (2012–2024) to evaluate changes in the use of sustainable finance keywords before and after Moody’s 2019 acquisition of VE. Sentiment analysis and statistical tests identify a significant increase in the use of sustainability terms post-acquisition, reflecting normalization and diminished association with credit downgrades. Pre-acquisition, such terms conveyed heightened credit risk. Post-acquisition, they became routine, positively framed and less correlated with negative outcomes. The methodology includes textual analysis, sentiment scoring using the Loughran–McDonald dictionary, regression models, and Chow tests to assess structural breaks in keyword trends.FindingsThe study finds that Moody’s use of sustainable finance keywords increased significantly after its 2019 acquisition of VE, marking a shift toward routine integration of sustainability considerations. Pre-acquisition, such terms were rare, linked with downgrades and carried a negative tone, indicating heightened risk perceptions. Post-acquisition, keyword frequency rose, sentiment became less negative, and their association with downgrades diminished. This reflects a normalization of environmental, social and governance (ESG) factors in credit assessments, driven by Moody’s enhanced expertise, market trends and regulatory pressures. The findings highlight the evolving role of ESG in credit evaluations, influencing perceptions of credit risk and capital allocation.Practical implicationsThe study highlights the transformative role of strategic acquisitions in integrating ESG considerations into financial assessments. For credit rating agencies, it underscores the importance of aligning methodologies with evolving market demands and regulatory expectations. Policymakers can leverage these findings to promote standardized ESG reporting and encourage sustainable finance integration. Investors benefit from more nuanced credit assessments that incorporate ESG factors, enabling informed capital allocation aligned with sustainability goals. For corporations, the normalization of ESG considerations incentivizes improved sustainability practices to achieve favourable ratings, influencing access to capital and financial strategies while fostering a more resilient global financial system.Originality/valueThis study provides novel insights into how ESG considerations evolve within credit rating methodologies following strategic acquisitions. By analysing over 24,000 Moody’s reports, it is the first to demonstrate a significant shift in the frequency, sentiment and impact of sustainable finance keywords post-acquisition. Unlike prior research, which focuses on ESG’s quantitative effects, this study explores narrative changes, revealing how sustainability factors transitioned from exceptional risks to routine considerations. It highlights the broader systemic forces – acquisitions, market demands, and regulation – that drive ESG integration, offering valuable contributions to academic discourse, financial market practices and sustainable investment strategies.