Search results

The purpose of this paper is to review and discuss the literature focusing on defining and measuring sentiments so as to understand their role in stock market behavior.

Critical review of the literature by analyzing myriad scholarly articles. The study is based on an analysis of 81 scholarly articles to critically analyze the approach toward defining and measuring market sentiments. The articles have been examined to identify and critique different classification of sentiment measures. A discussion is built to scrutinize the sentiment measures under the purview of theoretical underpinnings of the investor sentiment theory as well.

With more than five decades of research, the sentiment construct in finance literature is still ill-defined. Myriad empirical proxies of sentiment measures have led to conflicting results. The sentiment construct defined in financial theories needs to be revisited from the lens of sentiments defined in psychology.

The study is limited to analyzing the role of individual and institutional sentiments in equity markets. There is a need to explore sentiments with respect to different investment styles and strategies along with the type of investors.

Developing a suitable sentiment proxy can result in devising profitable trading strategies for investors. Understanding factors driving investor sentiments will help regulators to become more proactive and frame better policies.

This paper has leveraged psychology literature to highlight the limitations in development of sentiment construct in finance literature. By identifying stylized facts from reviewing the empirical literature, it highlights areas for future research.

The purpose of this study is to investigate whether the direct and indirect sentiment measures are similar in explaining mutual fund performance.

The authors examine the role of direct and indirect sentiment measures on fund performance in two scenarios. One is when a sentiment measure is added to market models, and the other is when it used independently. Also, the authors propose a system science theory to explain the findings.

The authors find that both direct and indirect sentiment measures are integral to the benchmark models to explain fund performance. However, while the explanatory power of the direct sentiment index is robust when used independently or collectively, the indirect sentiment measures can explain fund performance only when used along with other market factors.

Given the number of sentiment measures, it is critical to determine whether these measures contain the same information of sentiment. This paper represents the first study on this topic.

The research community currently employs four very different versions of the social network concept: A social network is seen as a set of socially constructed role relations (e.g., friends, business partners), a set of interpersonal sentiments (e.g., liking, trust), a pattern of behavioral social interaction (e.g., conversations, citations), or an opportunity structure for exchange. Researchers conventionally assume these conceptualizations are interchangeable as social ties, and some employ composite measures that aim to capture more than one dimension. Even so, important discrepancies often appear for non-ties (as dyads where a specific role relation or sentiment is not reported, a specific form of interaction is not observed, or exchange is not possible).

Investigating the interplay across the four definitions is a step toward developing scope conditions for generalization and application of theory across these domains.

This step is timely because emerging tools of computational social science – wearable sensors, logs of telecommunication, online exchange, or other interaction – now allow us to observe the fine-grained dynamics of interaction over time. Combined with cutting-edge methods for analysis, these lenses allow us to move beyond reified notions of social ties (and non-ties) and instead directly observe and analyze the dynamic and structural interdependencies of social interaction behavior.

This unprecedented opportunity invites us to refashion dynamic structural theories of exchange that advance “beyond networks” to unify previously disjoint research streams on relationships, interaction, and opportunity structures.

This study examines the effect of firm-level investor sentiment on a firm's level of financial distress.

The authors use Bloomberg's firm-level, daily investor sentiment scores derived from firm-level news and Twitter content in a beta-regression model to explain the variability in a firm's financial distress.

The results indicate that improvements (deterioration) in investor sentiment derived from both news articles and Twitter content lead to a decrease (increase) in the average firm's financial distress level. We also find that the effect of sentiment derived from Twitter on a firm's financial distress is significantly stronger than the sentiment derived from news articles.

Our proxy for financial distress is Bloomberg's financial distress measures, which may be an imperfect measure of financial distress. Our results have important implications for market participants in assessing the determinants of financial distress.

Our sample period covers four years (2015–2019), which is determined by Bloomberg sentiment data availability.

Market participants are increasingly using social media to express views on firms and seek information that might be used to determine a firm's level of financial distress. Our study links investor sentiment derived from social media (Twitter) and traditional news articles to financial distress.

By examining the relationship between a firm's sentiment and its financial distress, this paper advances our understanding of the factors that drive a firm's financial distress. To our knowledge, this is the first study to link US firms' investor sentiment derived from firm-level news and Twitter content to a firm's financial distress.

This study creates a measure of investor sentiment directly from retail trader activity to identify misvaluation and to examine the link between sentiment and subsequent returns.

Using investor reports from a large discount brokerage that include measures of activity such as net buying, net new accounts and net new assets, this study creates a measure of retail trader sentiment using principal components. This study examines the relation between sentiment and returns through conditional mean and regression analyses.

Retail sentiment activity coincides with aggregate Google Trends search data and firms with the greatest sensitivity to retail sentiment tend to be small, young and volatile. Periods of high retail sentiment precede poor subsequent market returns. Cross-sectional results detail the strongest impact on subsequent returns within difficult to value or difficult to arbitrage firms.

This study links a rich measure of retail trader activity to subsequent market and cross-sectional returns. These results deepen our understanding of noise trader risk and aggregate investor sentiment.

The purpose of this paper is to examine the effect of firm-level investor sentiment on a firm's share liquidity.

The authors use Bloomberg's firm-level, daily investor sentiment scores derived from firm-level news and Twitter content in a regression model to explain the variability in a firm's share liquidity.

The results indicate that improvements (deterioration) in investor sentiment derived solely from Twitter content lead to a decrease (increase) in the average firm's share liquidity. Results, although not as strong, are opposite for investor sentiment derived solely from news articles: improvements (deterioration) in news sentiment leads to an increase (decrease) in the average firm's share liquidity.

The proxy for share liquidity is the bid-ask spread, which may be an imperfect measure of liquidity. The Amihud illiquidity measure was used as an alternative proxy and yield similar results. The results have important implications for investors in assessing the determinants of share liquidity.

The sample period covers four years (2015–2018), which is determined by the availability of the Bloomberg sentiment data.

Investors increasing use of social media to express views on particular stocks and seek information that might be used in the investment decision-making process. The study links investor sentiment derived from social media (Twitter) to share liquidity.

By examining the relationship between a firm's sentiment and the firm's share liquidity, this paper advances the authors' understanding of the factors that drive a firm's share liquidity. To the authors' knowledge, this is the first study to link investor sentiment derived from firm-level news and Twitter content to a firm's share liquidity.

A composite sentiment index (CSI) from quantitative proxy sentiment indicators is likely to be a lag sentiment measure as it reflects only the information absorbed in the market. Information theories and behavioral finance research suggest that market prices may not adjust to all the available information at a point in time. This study hypothesizes that the sentiment from the unincorporated information may provide possible market leads. Thus, this paper aims to discuss a method to identify the un-incorporated qualitative Sentiment from information unadjusted in the market price to test whether sentiment polarity from the information can impact stock returns. Factoring market sentiment extracted from unincorporated information (residual sentiment or sentiment backlog) in CSI is an essential step for developing an integrated sentiment index to explain deviation in asset prices from their intrinsic value. Identifying the unincorporated Sentiment also helps in text analytics to distinguish between current and future market sentiment.

Initially, this study collects the news from various textual sources and runs the NVivo tool to compute the corpus data’s sentiment polarity. Subsequently, using the predictability horizon technique, this paper mines the unincorporated component of the news’s sentiment polarity. This study regresses three months’ sentiment polarity (the current period and its lags for two months) on the NIFTY50 index of the National Stock Exchange of India. If the three-month lags are significant, it indicates that news sentiment from the three months is unabsorbed and is likely to impact the future NIFTY50 index. The sentiment is also conditionally tested for firm size, volatility and specific industry sector-dependence. This paper discusses the implications of the results.

Based on information theories and empirical findings, the paper demonstrates that it is possible to identify unincorporated information and extract the sentiment polarity to predict future market direction. The sentiment polarity variables are significant for the current period and two-month lags. The magnitude of the sentiment polarity coefficient has decreased from the current period to lag one and lag two. This study finds that the unabsorbed component or backlog of news consisted of mainly negative market news or unconfirmed news of the previous period, as illustrated in Tables 1 and 2 and Figure 2. The findings on unadjusted news effects vary with firm size, volatility and sectoral indices as depicted in Figures 3, 4, 5 and 6.

The related literature on sentiment index describes top-down/ bottom-up models using quantitative proxy sentiment indicators and natural language processing (NLP)/machine learning approaches to compute the sentiment from qualitative information to explain variance in market returns. NLP approaches use current period sentiment to understand market trends ignoring the unadjusted sentiment carried from the previous period. The underlying assumption here is that the market adjusts to all available information instantly, which is proved false in various empirical studies backed by information theories. The paper discusses a novel approach to identify and extract sentiment from unincorporated information, which is a critical sentiment measure for developing a holistic sentiment index, both in text analytics and in top-down quantitative models. Practitioners may use the methodology in the algorithmic trading models and conduct stock market research.

Although the effects of both news sentiment and expectations on price in financial markets have now been extensively demonstrated, the jointness that these predictors can have in their effects on price has not been well-defined. Investigating causal ordering in their effects on price can further our understanding of both direct and indirect effects in their relationship to market price.

We use autoregressive distributed lag (ARDL) methodology to examine the relationship between agent expectations and news sentiment in predicting price in a financial market. The ARDL estimation is supplemented by Grainger causality testing.

In the ARDL models we implement, measures of expectations and news sentiment and their lags were confirmed to be significantly related to market price in separate estimates. Our results further indicate that in models of relationships between these predictors, news sentiment is a significant predictor of agent expectations, but agent expectations are not significant predictors of news sentiment. Granger-causality estimates confirmed the causal inferences from ARDL results.

Taken together, the results extend our understanding of the dynamics of expectations and sentiment as exogenous information sources that relate to price in financial markets. They suggest that the extensively cited predictor of news sentiment can have both a direct effect on market price and an indirect effect on price through agent expectations.

Even traditional financial management firms now commonly track behavioral measures of expectations and market sentiment. More complete understanding of the relationship between these predictors of market price can further their representation in predictive models.

This article extends the frequently reported bivariate relationship of expectations and sentiment to market price to examine jointness in the relationship between these variables in predicting price. Inference from ARDL estimates is supported by Grainger-causality estimates.

This study examines the impact of social media investor sentiment on the stock market performance through qualitative and quantitative proxies.

The authors use a sample of daily stock performance related to S&P 500 Index for the period from December 18, 2017, to December 18, 2018. The social media investor sentiment was assessed through qualitative and quantitative proxies. For qualitative proxies, the study relies on three social media resources”: Twitter, Trump Twitter account and StockTwits. The authors proposed 3 methods to reflect investor sentiment. For quantitative proxies, the number of daily messages published from Trump Twitter account and StockTwits is considered as a signal of investor sentiment. For regression model, the study adopts the autoregressive distributed lagged to determine the relationships between the nonstationary series.

Empirical findings provide evidence that quantitative measures of investor sentiment have significant effects on S&P’500 performances. The authors find that Trump's tweets should be interpreted with caution. The results also show that the number of Trump's tweets on t−1 day have a positive effect on performance on day t.

Social media sentiment contains information for predicting stock returns and transaction activity. Since, the arrival of new information in capital markets triggers investor sentiment on social media.

This study investigates the investors’ sentiment through social media and explores quantitative and qualitative measures. The amount of information on social media reflects more the investor sentiment than content analysis measures.

The peer review history for this article is available at: https://publons.com/publon/10.1108/IJSE-12-2022-0818

The authors compare sentiment level with sentiment shock from different angles to determine which measure better captures the relationship between sentiment and stock returns.

This paper examines the relationship between investor sentiment and contemporaneous stock returns. It also proposes a model of systems science to explain the empirical findings.

The authors find that sentiment shock has a higher explanatory power on stock returns than sentiment itself, and sentiment shock beta exhibits a much higher statistical significance than sentiment beta. Compared with sentiment level, sentiment shock has a more robust linkage to the market factors and the sentiment shock is more responsive to stock returns.

This is the first study to compare sentiment level and sentiment shock. It concludes that sentiment shock is a better indicator of the relationship between investor sentiment and contemporary stock returns.