Sentiment Index Research Articles

Impactful messaging: Elite sentiment in Chinese new energy vehicle vs machine learning perspective

Elite messaging is sensitive to framing decisions and shaping public sentiment. This study examines the influence of elite sentiment on the pricing of the stock index of China's new energy vehicles (NEVs). To construct elite sentiment indexes for China's NEVs, data from a highly active online elite forum were collected and a series of machine learning models were utilized to generate predictions. The results demonstrate that the elite sentiment index has a pronounced predictive impact on stock index prices. Furthermore, incorporating the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) method can further enhance the baseline model's predictive ability.

The impact of consumer credit on the stability of the banking system: analysis of economic and psychological factors

The purpose of the research is to determine the impact of consumer lending on bank sustainability and to assess its dependence on economic indicators.Methodology. The research methodology is based on qualitative and quantitative approaches. The methods of regression analysis, transformation of analytical data mapping, analysis and synthesis were used in the article.Originality / value of the research lies in the fact that the authors have determined the dependence of consumer lending on both macroeconomic and psychological factors on the basis of calculations.Findings. The authors, through econometric calculations, obtained the consumer function as the main indicator of aggregate demand, which affects economic growth. They revealed a low dependence of consumer credit on: interest rate, average monthly income of the population and inflation. The result of multiple regression showed that macroeconomic factors explained only 43,3 % of consumer credit growth. Along with this, the authors identified a significant impact of consumer credit on imports and gross value added of trade.Based on qualitative analysis, the authors calculated the Consumer Sentiment Index (CCI), based on a survey of 128 respondents. The survey found that consumers are optimistic and highly susceptible to psychological instruments of influence, such as marketing and promotions. Also, the authors believe that the perception of debt burden and credit availability influenced the dramatic growth of this banking product.Arguing their assumptions by citing contemporary research by economists, the authors determined that changes in demand for consumer credit are determined more by psychological factors than economic ones. The increase in household final expenditures was due to an increase in consumer credit, which stimulated aggregate demand in the short run and thereby influenced economic growth. At the same time, in recent years in Kazakhstan, lending to households has outpaced lending to industries and the ratio of household debt to GDP has been increasing. This trend may lead in the long run to bank defaults and economic recession in the country. Based on this, the authors have proposed some recommendations to address these problems.

EMD-BiLSTM Stock Price Trend Forecasting Model based on Investor Sentiment

The movement of stock prices is the focus of investors' attention in the stock market, so stock price trend prediction has always been a hot topic in quantitative investment research. Traditional machine learning forecasting models are difficult to process nonlinear, high-frequency, high-noise stock price time series, which makes the prediction accuracy of stock price trends low. In order to improve the prediction accuracy, according to the temporal characteristics of stock price data, it is proposed to use a combination of empirical mode decomposition (EMD), investor sentiment and two-way long short-term memory neural network to predict the rise and fall of stock prices. Firstly, the empirical mode decomposition algorithm is used to extract the characteristics of the stock price time series on different time scales, and the investor sentiment indicators of the text from the close of the previous stock trading day to the opening of the next trading day are extracted by constructing a financial sentiment dictionary, and finally the EMD-BiLSTM model is used to predict the rise and fall of the next index trading day. Experiments on the dataset of stock price series show that the optimized BiLSTM model has strong predictive ability for the trend of consumer sector indexes.

Don't miss out on NFTs?! A sentiment-based analysis of the early NFT market

This study investigates the impact of Twitter sentiment on the Non-Fungible Token (NFT) market. Using a dataset of over 5 million English-language tweets on NFTs, we calculate a daily sentiment index and link it to NFT sales and trading volume. Applying wavelet analysis and DCC-GJR-GARCH models, we analyze the NFT market, characterized by multiple bubbles and high volatility. The findings reveal Twitter's significance as a primary source of information for a broad audience. Moreover, the study contributes to the literature by examining the role of Twitter sentiment in the NFT market's development. Additionally, the study indicates weak links between established cryptocurrencies and the NFT market. Based on our findings, we recommend that traders and policymakers use social media activities to monitor new digital markets.

Is there a CSI-leverage nexus?

We examine how various measures of consumer sentiment index (CSI) affect firms' debt policy decisions. Using U.S. firm-level quarterly data from 1993 to 2017, we provide a strong positive relationship between CSI measures and corporate debt policy, implying that firms use external borrowing during a positive economic outlook and reap the tax-shield benefit. We also find that improved household optimism over financial and business sentiments leads to future household consumption. The CSI-leverage nexus is moderated by the state of firms' financial condition, reputation, and profitability. Importantly, our results are robust to sub-sample analysis, firm-level and macroeconomic controls, econometric specifications, alternative measures of sentiment including Shiller's cyclically adjusted price-earnings ratio (i.e., CAPE_SH), Baker and Wurgler (2006)’s stock market sentiment index (i.e., SENT_BW) and search-based uncertainty measure such as FEARS (i.e., Financial and Economic Attitudes Revealed by Search) index of Da, Engelberg, and Gao (2015).

The Impact of Sentiment Indices on the Stock Exchange—The Connections between Quantitative Sentiment Indicators, Technical Analysis, and Stock Market

The stock market represents one of the most complex mechanisms in the financial world. It can be seen as a living being with complex ways to enact, interact, evolve, defend, and respond to various stimuli. Technical analysis is one of the most complex techniques based on financial data’s graphical aspects. News sentiment indices are very complex and highlight another important part of behavioral finance. In this study, we propose an integrated approach in order to determine the correlation between news sentiment indices, the stock market, and technical analysis. The research methodology focuses on the stock market’s practical and quantitative aspects. In this sense, we have used the graphical representation of technical analysis and econometric modeling techniques such as VAR and Bayesian VAR. The results of the empirical modeling techniques and analysis reveal some important connections between the stock market and news sentiment indices on the US stock market. The conclusions of this study highlight a strong connection between news sentiment indices, technical analysis, and the stock market which suggests that the behavioral finance aspect is a very important aspect in the analysis of the stock market.

The relationship between social media sentiment and house prices in China: Evidence from text mining and wavelet analysis

This study explores the dynamic relationship between house prices and social media sentiment in China, a housing sentiment index is constructed from social media reviews based on natural language processing techniques, and wavelet analysis is used to investigate the causal correlations from a new time-frequency perspective. We find that the sentiment index is negatively correlated with the fluctuation of house prices during the whole period, house prices and sentiment have bidirectional causality at the long-term timescale, and house prices causally affect sentiment at the short-term timescale. Furthermore, the sentiment significantly affects house prices in third-tier cities and the western regions.

Dynamic connectedness between global commodity sectors, news sentiment, and sub-Saharan African equities

We examine the static and time-varying spillovers between global commodity sectors, economic news sentiment, and sub-Saharan African (SSA) equities. In a time-varying parameter vector autoregressions spillover connectedness approach, we find that: (i) there are significant spillovers between SSA equities, economic news sentiment, and global commodity sectors; (ii) the connectedness evolves across different times and is amplified by market crises; (iii) there are suitable market-specific SSA equities that bear safe-haven and hedge attributes for global commodity investors; (iv) agriculture is the main source of return and volatility spillovers; and (v) economic news sentiment index is a net transmitter of spillover during systemic crisis periods such as GFC and the COVID-19 pandemic. Our findings have notable implications for international investors, regulators, and practitioners.

Developing central bank digital currencies: a reality check during cryptocurrency euphoria

Using estimated sentiment indices based on CBDC-related social media posts, and testing for the effects of regulatory-related announcements upon blockchain and cryptocurrency-related funds, this research presents two key findings: first, the continued evolution of the pricing structures of digital finance products to respond to such perceived threats constitutes a further evolutionary point in the product's life-cycle. However, secondly, the very fact that returns fall while volatility increases, indicates a largely negative market response to the threat of potential external regulation of cryptocurrencies in the future. The nature of this negative response validates concerns that anonymity continues to be a central attractive feature for cryptocurrency stakeholders, further verifying the necessity for third-party oversight.

Retail Demand Forecasting Using Neural Networks and Macroeconomic Variables

With the growing competition among firms in the globalized corporate environment and considering the complexity of demand forecasting approaches, there has been a large literature on retail demand forecasting utilizing various approaches. However, the current literature largely relies on micro variables as inputs, thereby ignoring the influence of macroeconomic conditions on households’ demand for retail products. In this study, I incorporate external macroeconomic variables such as Consumer Price Index (CPI), Consumer Sentiment Index (ICS), and unemployment rate along with time series data of retail products’ sales to train a Long Short-Term Memory (LSTM) model for predicting future demand. The inclusion of macroeconomic conditions in the predictive model provides greater explanatory power. As anticipated, the developed model, including this external macroeconomic information, outperforms the model developed without this macroeconomic information, thereby demonstrating strong potential for industry application with improved forecasting capability.

Dispersion in news sentiment and corporate bond returns

We construct a news sentiment index at the firm level by using textual analysis of news articles and find that dispersion in news sentiment is a significant predictor of corporate bond returns. Bonds of firms with high dispersion in news sentiment have a highly significant average return of 7.38 percent. A portfolio that longs bonds with high dispersion in news sentiments and shorts bonds with low dispersion earns an average biweekly return of 8.53 percent. This finding is in line with an argument that dispersion in news sentiment is a proxy for future cash flow uncertainty.

Consumer Sentiment in the United States and the Impact of Mental Disorders on Consumer Behavior—Time Trends and Persistence Analysis

This paper analyzes the stochastic properties in clinical disorders to understand how they have manifested in consumer sentiment in the USA since 1990. The results obtained via fractional integration methodologies exhibit a high degree of persistence, finding non-mean reversion behavior in all of the time series analyzed, except for depressive disorder. Using a causality test, we find that mental and substance use disorders, anxiety disorder, schizophrenia, and alcohol use disorder influence consumer sentiment. Focusing on the cointegrating part, we conclude that an increase in the previously cited mental disorders produces a decrease in the Consumer Sentiment Index.

Impact of Investor Sentiment on Contemporaneous and Future Equity Returns in Emerging Markets

The study is aimed to investigate the impact of investor sentiment on returns of selected emerging equity markets i.e., Brazil, India, China, Russia, Indonesia, and Pakistan using non-linear predictive regression analysis. Principal Component Analysis is used to generate investor sentiment index. Investor sentiment has a significant impact on current market returns and this influence is continued in the short run in most of the sample countries. However, the impact of sentiment is not much pronounced in the longer run. Therefore, decision-makers cannot ignore sentiments in at least the short run. The findings of this study may be helpful for investors to better understand the market trends under the influence of sentiments and portfolio managers and risk professionals can devise their strategies accordingly. JEL: G02, G15, G24, M41

Textual analysis and detection of financial fraud: Evidence from Chinese manufacturing firms

With the increasing complexity of financial statement manipulation, relying solely on quantitative financial data may not effectively detect financial fraud. While textual analysis can provide additional insight, little research has been conducted on its multiple dimensions. Using 579 listed Chinese manufacturing firms in 2020, we select readability, forward-looking, similarity, matching degree, and positive and negative sentiment indicators from textual language structure, quality, and expression of management discussion and analysis texts, in combination with financial indicators, to detect financial fraud. Our findings indicate that fraudulent firms tend to be overly cautious in their financial reporting, express fewer positive sentiments, and conceal financial fraud by increasing the complexity of their annual reports and using more degree adverbs to modify forward-looking information. This study also highlights the importance of considering textual language expression in detecting financial fraud in state-owned and non-state-owned firms.

Investor Sentiment, Cross-sectional Stock Returns, and Short-Sales: Evidence From Korea

Purpose: This study investigates the return co-movements associated with investor sentiment shifts in the cross-sections under a setting where market-wide sentiment interacts with short-sale impediments.
 Design/methodology/approach: This study estimates the return sensitivity to market sentiment changes (sentiment beta) for each characteristic's portfolio by regressing the return of each quintile portfolio and various high-minus-low portfolios on the sentiment changes index. It examines whether these cross-sectionally different return co-movement patterns is more prevalent during good- than bad-sentiment periods by performing the same regression separately for the good- and bad-sentiment periods.
 Findings: The result shows that the returns of speculative stocks tend to co-move more strongly with sentiment changes than those of stable stocks, in the sense that speculative stocks have higher sentiment betas than stable stocks. The cross-sectional pattern in return co-movements becomes more pronounced during the good-sentiment period but disappears during the bad-sentiment periods.
 Research limitations/implications: This study elucidates the return co-movement behavior associated with investor sentiment changes under a setting where market-wide sentiment interacts with short-sale impediments. However, analyzing the relationship between the investment sentiment index and the short-selling activities is reserved for future research.
 Originality/value: The results provide important implications for investment strategies using investor sentiment in practice, and several suggestions for establishing investor protection policies in the highly individual-crowded market. This study will contribute to enhancing the stock market efficiency and price discovery.

The impact of African swine fever news sentiment on the Korean meat market.

Our study analyzed the impact of African swine fever (ASF) news on the Korean meat market using sentiment analysis. We applied a neural network language model (NNLM) to generate a sentiment index indicating whether the news had a positive or negative impact on consumer expectations. We analyzed 24,143 news articles to estimate the impulse responses of meat price variables to sentiment shocks. Our study contributes significantly to agricultural economics as it applies NNLM to generate a sentiment index. The empirical results indicated that ASF news sentiment has a substantial impact on meat prices in Korea, and there is evidence of substitution effects among different types of meat. ASF news has a positive impact on the price of pork, negative effects on beef and chicken prices, and a greater impact on the price of chicken than beef. The findings imply that the effect of ASF news on demand outweighs its impact on supply in the pork market, whereas the effect on supply surpasses the effect on demand in the beef and chicken market. We believe our methods and results will inspire discussions among applied economists studying consumer behavior in this specific market and could encourage the application of big data analysis to the agricultural economy.

Data mining of social media for urban resilience study: A case of rainstorm in Xi'an

Disasters throughout the world have highlighted the urgent need for studies on urban resilience capacities. Individuals' collective responses to disasters provide significant insights into their capacity for catastrophes adaptation and an invaluable perspective for demonstrating urban resilience. Previous research on urban resilience has rarely assessed resilience on an urban scale based on individuals' collective responses. As social media data are both a personal expression of users and a subjective reflection of their surroundings, we developed a social media-based data mining method to reflect urban resilience. We examined the urban resilience with the case of a rainstorm in Xi'an on July 24, 2016. The temporal patterns of public behavioral and emotional dimensions were chosen to examine how the city tolerated the disruption and returned to normal conditions. The public's behavioral reactions peaked during the rainstorm and lasted for a long time, revealing a delayed and hysteretic response. After 33 h, the holistic sentiment index repaired from 0.412 to the baseline standard of 0.713, indicating relatively feeble urban resilience to such rainstorm disasters. Our research revealed the viability of using social media to quantify a city's resilience, which is significantly beneficial in supplementing previous research.

A central bankers’ sentiment index of global financial cycle

We propose a novel Central Bankers’ Sentiment Index (CBSI) to measure the global financial cycle (GFC) using text analysis on 6,514 central bankers’ speeches. We find that the CBSI outperforms the conventional measures in capturing the GFC by tracing historical events and identifying turning points. Our empirical evidence further demonstrates the comparative advantages of CBSI in reflecting the capacities of central banks to aggregate financial system information and intervene in financial market operations. Our findings provide new insights into a more accurate measurement of the GFC and also contribute to a better understanding of the impact of central bank speeches.

Lost in translation. When sentiment metrics for one market are derived from two different languages

We compare two text-based proxies for the sentiment of investors in the Japanese market. Both proxies are constructed by Thomson Reuters using the same algorithm, and the only difference between them is that the first proxy is derived using only Japanese language items; the second is derived from English language items. The correlation between the proxies is low and this suggests that they measure different aggregate affective states. The English-language sentiment proxy is found to have a positive and statistically significant association with Japanese returns before a key date in the history of Abenomics relating to quantitative easing (April 2013). After this date, only the Japanese-language proxy has a positive and statistically significant association with the market. Studies of sentiment are predominantly based on proxies derived from English language sources; our results contribute to this nascent research field and suggest that not using both the English and local-language sources when constructing text-based sentiment indices may lead to misleading or mistaken inferences.

Investors’ Irrational Sentiment and Stock Market Returns: A Quantile Regression Approach Using Indian Data

Studying sentiment is crucial for investors and portfolio managers to determine whether sentiment can be used as information to make profits. This study examines the relationship between irrational sentiment among investors and excess returns in the Indian stock market using monthly data from July 2001 to December 2019. The study constructs a composite sentiment index that includes condensed information from 10 variables. The empirical analysis reveals that the influence of irrational sentiment among investors on excess returns in the stock market is not uniform across quantiles. Specifically, our results indicate that the irrational sentiment index has information regarding contemporaneous (future) variation in excess returns in upper (all) quantiles. Results also suggest that the predictive ability of irrational sentiment is enhanced when market conditions are right. We also decompose the sentiment into positive and negative irrational sentiment and find an asymmetrical impact in upper quantiles but lost at lower quantiles.