Accounting for investor sentiment in news and disclosures

Abstract

Purpose This study aims to develop two sentiment indices sourced from news stories and corporate disclosures of the firms in the National Stock Exchange NIFTY 50 Index by extracting sentiment polarity. Subsequently, the two indices would be compared for the predictive accuracy of the stock market and stock returns during the post-digitization period 2011–2018. Based on the findings this paper suggests various options for financial strategy. Design/methodology/approach The news- and disclosure-based sentiment indices are developed using sentiment polarity extracted from qualitative content from news and corporate disclosures, respectively, using qualitative analysis tool “N-Vivo.” The indices developed are compared for stock market predictability using quantitative regression techniques. Thus, the study is conducted using both qualitative data and tools and quantitative techniques. Findings This study shows that the investor is more magnetized to news than towards corporate disclosures though disclosures contain both qualitative as well as quantitative information on the fundamentals of a firm. This study is extended to sectoral indices, and the results show that specific sectoral news impacts sectoral indices intensely over market news. It is found that the market discounts information in disclosures prior to its release. As disclosures in quarterly statements are delayed information input, firms can use voluntary disclosures to reduce the communication gap with investors by using the internet. Managers would do so only when the stock price is undervalued and tend to ignore the market and the shareholder in other cases. Otherwise, disclosure sentiment attracts only long horizon traders. Practical implications Finance managers need to improve disclosure dependence on investors by innovative disclosure methodologies irrespective of the ruling market price. In this context, future studies on investor sentiment would be interesting as they need to capture man–machine interactions reflected in market sentiment showing the interplay of human biases with machine-driven decisions. The findings would be useful in developing the financial strategy for protecting firm value. Originality/value This study is unique in providing a comparative analysis of sentiment extracted from news and corporate disclosures for explaining the stock market direction and stock returns and contributes to the behavioral finance literature.

Mixed frequency composite indicators for measuring public sentiment in the EU

“…A similar advantage can also be gained by encompassing the media sentiment in the confidence measure, as in the MF-CCI + . The critical reading of the economic phenomena given by newspapers and news agencies could offer an alternative but concurrent viewpoint that have to be taken into account by policymakers and politicians at large, primarily because the opinions of media often reflect the official and unofficial positions of relevant interest groups and intermediary bodies (Binderkrantz et al 2017;Dür 2019;Eachempati and Srivastava 2022). On the other hand, as stressed above, the use of confidence indicators as significant leading indicators in GDP predictions has been recently debated by scholars and practitioners, highlighting how this kind of data could be employed within everyday economic analysis in a "nowcasting" perspective.…”

Mainstreaming environmental, social and governance integration in investment practices in South Africa: A proposed framework

“…Yet Fama (1998) also asserted that anomalies tend to disappear over time, and hence markets tend to become more efficient over longer-term periods. Research has shown that share market participants often make material mistakes and frequently over-react or under-react to news of corporate success or failure (Eachempati & Srivastava 2021;Gang, Qian & Xu 2019). The proponents of behavioural finance theory therefore argue that inefficient markets are the norm and that efficient markets are the exception (Vasileiou 2021;Willett 2022;Yen & Lee 2008).…”