Sentiment Factors Research Articles

ANALYZING SOCIAL MEDIA SENTIMENT TOWARD SPECIFIC COMMODITIES FOR FORECASTING PRICE MOVEMENTS IN COMMODITY MARKETS

This study adopts a systematic literature review to analyze social media sentiment towards specific commodities to enhance the accuracy of price movement forecasts in commodity markets. Drawing from the field of applied mathematics, the research gathered literature from Scopus, DOAJ, and Google Scholar databases, covering publications from 2014 to 2024. A rigorous search strategy yielded 66 journal articles, with 30 being selected for their close relevance to keywords such as "social media sentiment," "commodity markets," and "price forecasting." Results indicate that social media sentiment significantly influences commodity prices, with particular variations based on commodity type and geographical context. Specific sentiment factors—especially intensity, polarity, and timing—were found to have a pronounced impact on price dynamics, with sentiment polarity being particularly influential in volatile markets. Additionally, advanced analytical methods, like Bayesian Dynamic Linear Models and LSTM neural networks, enhance predictive accuracy when applied to sentiment analysis in this context. These findings underscore the value of social media sentiment in refining forecasting models, while also highlighting gaps in understanding regional sentiment variations and their effects on different commodity types. By synthesizing these insights, this study emphasizes the importance of considering social media sentiment for more accurate price predictions and identifies key areas for future research to explore the multifaceted impacts of sentiment in commodity markets.

Sentiments in the cryptocurrency market: an in-depth analysis of influential factors applying ISM-MICMAC and AHP

PurposeCryptocurrency, an emerging asset class, is a virtual form of currency that uses cryptography for security and operates on decentralised networks based on blockchain technology. It offers both challenges and opportunities for investors, particularly in terms of diversification, risk management and potential returns. Considering this, the present study attempts to investigate the sentimental factors influencing cryptocurrency while unravelling the intricate interplay among these factors.Design/methodology/approachTo achieve this, interpretive structure modelling (ISM) identifies the hierarchical model of critical sentimental factors, while Cross-Impact Matrix Multiplication Applied to Classification (MICMAC) explores their dependency and driving power. Analytic hierarchy process (AHP) is adopted to rank the drivers.FindingsFindings reveal that the pandemic, war, religiosity and economic uncertainty are top-level factors dominantly shaping cryptocurrency trends. Simultaneously, Google Search Trends and Herding emerge as the most dependent factors, influenced by sentiments that emerged from other factors.Practical implicationsThe study unpacks implications, acknowledges limitations and proposes avenues for future research.Originality/valueBy exploring the interactive interrelationships among identified sentimental factors through ISM-MICMAC analysis and ranking via the AHP, this paper will have a great influence while contributing towards this evolving field.

A comparative study of bitcoin price prediction during pre-Covid19 and whilst-Covid19 period using time series and machine learning models

Investment in cryptocurrencies has garnered substantial attention in the recent past as the prices for these digital currencies started recording all-time highs. While there are numerous contenders in the cryptocurrency market, bitcoin has emerged to be the most popular and sought after digital currency. Despite its popularity, the theoretical understanding of the value of this cryptocurrency is still limited. Hence this study aims to find out the significant predictors of the bitcoin price and build a machine-learning based model to evaluate and predict the complex phenomenon of bitcoin price. Here we contribute to the extant literature by searching for the potential contributors of bitcoin prices ranging from fundamental, macroeconomic, financial, speculative, and technical sources to the most marked event of 2020 i.e., Covid19 pandemic. For this purpose, we have used state-of-the-art machine learning, deep learning, and statistical time-series models (univariate and multivariate) to forecast bitcoin price. The study revealed that deep learning models performed almost at par with Random Forest model for both pre- and whilst-Covid19 era. Traditional time-series models, namely VAR and VECM gave the most consistent performance within acceptable margins for both pre- and whilst-Covid era. We have also found that macroeconomic factors play an important role in determining bitcoin price formulation process during both periods, while mining difficulty and market sentiment factors gain more importance during pre-Covid period. In addition, number of covid cases is also found to be a significant factor for the prediction of bitcoin price during whilst-Covid period.

Decrypting Metaverse crypto Market: A nonlinear analysis of investor sentiment

This study aims to investigate the role of investor sentiment in the emerging metaverse market, a novel entrepreneurship model. Empirical analyses are conducted through various causality tests to reveal the predictive power of investor sentiment on the price developments of the metaverse market. The Nonlinear Granger causality test indicates causal effects running from BTC (Bitcoin), GT (Google Trend), and FGI (Fear-Greed Index) to MVI (Metaverse Index). Further examination of these interactions through MS-VAR analysis reveals that under bear market regimes, both investor sentiment proxies (GT and FGI) and BTC have a statistically significant causal effect on the returns of MVI. This finding suggests that metaverse crypto market returns are substantially influenced by investor sentiment during periods of anxiety and turmoil, evident in steep bear markets, rather than during periods of tranquility and euphoria characteristic of bull markets. The results of the time-varying approach confirm this finding by indicating spikes in causal effects towards the end of 2021, during which a severe crash in cryptocurrency markets occurred. Overall, the causal links during market downturns may stem from the fear of missing out (FOMO) in retail investors, who mainly dominate the sentimental factors utilized in this study.

Research on Stock Price Prediction Model Based on Sentiment Factor and Multi-Core Bagging Algorithm

Stock price prediction model is one of the key research points in the quantitative financial field. With the rapid development of data acquisition and storage technology, the data of emotional factors based on nonlinear structure is increasing. How to combine this part of information with common European technology factors to increase the ability of the model is a problem that needs to be solved. Based on this, this paper proposes a method of uncertainty quantification that combines European-and non-European data. Specifically, on the one hand, the Gaussian process regression model is used to capture the unknown relationship between predictive variables and price; on the other hand, the multi-kernel learning technology based on European and non- European kernel functions is used to combine the effective information of European and non- European predictive variables. Moreover, the proposed method can weigh the bias of the prediction model against the variance by using the Bagging algorithm. The simulated data analysis shows that the proposed method not only improves the prediction performance of the European Gaussian process, but also works in the presence of irrelevant predictive variables. Finally, the actual data analysis shows that the proposed method has achieved good prediction results in the task of predicting the rise and fall of stock prices.

How green screening influences risk transmission among stock-bond indices: Insight into the dependence structure

Green characteristics are pivotal to risk transmission. We use environmental, social and governance (ESG) monetized accounting to measure green screening of stock-bond indices, and decompose risk transmission into comovement, contagion and hedging effects via a patched dependence structure model. Mechanisms through which green screening impacts three types of risk transmission effects are elucidated. Results indicate that high level of green screening is associated with significantly reduced comovement and contagion effects while hedging effects. Market sentiment factors such as interest rates, economic policy uncertainty and consumer confidence are identified as crucial channels. These results are further tested to be robust.

Iterative Deep Learning Approach to Active Portfolio Management with Sentiment Factors

Iterative Deep Learning Approach to Active Portfolio Management with Sentiment Factors

Integration of Modified Heston Model (Jump and Sentiment Factor) with Artificial Neural Networks (ANN) for Volatility Forecasting in Financial Markets

This study examined a new methodology for predicting volatility in financial markets by combining the modified Heston model (JSF) with Artificial Neural Networks (ANN). Forecasting volatility is essential in various financial contexts, such as risk management, option pricing and portfolio optimization. Although the traditional Heston model is effective but it does not account for additional elements like jumps and investor sentiment factors, which impact asset price dynamics. To address this limitation, the extended Heston model with jump and sentiment variables helped in enhancing its ability to capture complex market trends. We utilized ANN's predictive capabilities on the modified Heston model to improve volatility predictions using historical random market data from NASDAQ Composite. The ANN was trained on simulated data from the modified Heston model and assessed on its effectiveness in forecasting volatility. This approach helped in developing a forecasting hybrid semi-parametric model known as the modified-Heston-ANN model. Our findings revealed that this integrated approach outperformed the individual model, offering more precise and robust volatility predictions

MD&A tone and stock returns

This paper examines the impact of the MD&A (Management’s Discussion and Analysis) tone on cross-sectional stock returns in the unique context of Chinese A-share companies. We find a significant and positive relationship between the MD&A tone and Chinese stock returns after controlling for quantitative financial metrics. Our results suggest that the MD&A tone provides additional value that further enhances the information quality in MD&A disclosures and market responses. Furthermore, we observe that the textual tone effect is more pronounced for firms with lower institutional ownership, lower financial transparency, in less competitive market environments, and both the information factor and the sentiment factor of MD&A tone positively affect stock returns. Thus, we provide convincing evidence that MD&A narratives can serve as an effective communication tool to facilitate disclosure and mitigate information asymmetry between corporations and investors.

One crash, too many: Global uncertainty, sentiment factors and cryptocurrency market

Recent studies document that cryptocurrencies offer an alternative store of value, medium of exchange and can be used to hedge against currency and price fluctuations. However, the frequent collapse of the crypto-market undermines its safe-haven characteristics, as investors’ fear and anxiety could intensify market volatility and trigger a financial crisis. Motivated by the current global vicissitudes, this study examines the impact of uncertainty and sentiment factors on price behaviour of cryptocurrencies. To estimate our model, we used daily, low, high and closing price data for major crypto projects, from January 2018 to January 2023. We show that economic and political uncertainty factors significantly drive crypto prices. Furthermore, the interaction between sentiment dynamics as expressed by investors on different social platforms has a significant adverse effect on the returns of the cryptocurrency market, and the impact is more pronounced for tokens within the same ecosystem. Using the asymmetric GARCH-MIDAS model and TVP-VAR, we also demonstrate the existence of a significant contagion among tokens within the same ecosystem when bad (or good) news occurs. Considering the massive unprotected losses incurred by crypto investors during crises, our results provide important insights into how portfolio managers can effectively design investment strategies.

Portfolio Analysis of Multi-Factor Stock Selection Model Based on Investor Sentiment

Since 1990, China's securities market has been gradually improved. Domestic and foreign scholars have proposed many theoretically perfect investment models, but not many of them have been applied in practice because the market cannot satisfy the strict assumption conditions of the models. Foreign quantitative investment models are not necessarily effective in the Chinese market. Based on this, this paper takes the portfolio theory as a guide, uses the scoring method to construct a multi-factor stock selection model, and selects the relevant factors in combination with the characteristics of the Chinese market. Through empirical analyses, the validity of the multi-factor stock selection model with equal weight allocation in China's market is verified. In addition, this paper innovatively constructs an investor sentiment factor, combines the sentiment timing with the multi-factor stock selection model, optimises the timing on the basis of the multi-factor stock selection model, and proposes an investment strategy to obtain excess returns stably.

Impact of sentimental factors on stock portfolio returns –an empirical analysis

This study aims to introduce an integrated model for understanding the influence of various sentimental factors in conjunction with macroeconomic factors on portfolio returns across ten industry sectors within the US market. These sentimental factors are categorized into market-wide, consumer, and individual stock market factors to assess their impact on industry portfolio returns. Employing the Autoregressive Distributed Lag (ARDL) model, the study evaluates the effects of macroeconomic and sentimental factors on stock market portfolio returns. The findings reveal a negative relationship between short-term interest rates and portfolio returns in specific industry sectors like manufacturing, telecom, and wholesale/retail. The study finds a positive relationship between the Hi-tech sector's risk spread and portfolio returns. Market sentimental factors positively influence portfolio returns of durable, non-durable, utility, and other sectors. Individual sentimental factors negatively impact portfolio returns in hi-tech, utility, durable, energy, and other sectors. The stock market-related individual, sentimental factor of the number of IPOs has a positive impact on portfolio returns in the energy sector and a negative impact on portfolio returns in other sectors. Consumer sentimental factors are significant positive determinants for portfolio returns in durable, energy, telecom, health, and other sectors. Discounts on closed-end funds may provide vital fundamental information regarding lower future earnings for stocks in the durable and energy sectors. The study provides valuable insights for investors to optimize their portfolio strategies in response to macroeconomic and sentimental factors within specific industry sectors.

BIT COIN PRICE PREDICTION USING RF ALGORITHM WITH SENTIMENT FACTORS.

Sentiment analysis was included into a predictive model that was created for this project in order to forecast bitcoin closing values. The dataset contains daily Bitcoin data for January 2024, including sentiment scores computed for each day as well as opening, high, low, and closing values. The study assesses the predictive ability of the Random Forest regression model by using metrics like Mean Squared Error (MSE), Mean Absolute Error (MAE), and R-squared (R^2) score. The negative R^2 score indicates that, although the model may produce predictions, its overall fit to the data is still not ideal, implying that the model is unable to adequately represent the correlation between sentiment ratings and Bitcoin prices. Potential areas for development to increase the predicted accuracy of the model and gain a deeper understanding of how sentiment analysis affects Bitcoin prices include feature engineering, hyper parameter tuning, and model selection. Keywords: Cryptocurrency, Price Prediction, Machine learning, Random Forest

An affine term structure model with Fed chairs’ speeches

I analyze the impact of the sentiment expressed in speeches by Fed chairs on Treasury bond yields, using an affine term structure model. The speeches by Fed chairs are numerically evaluated to generate three sentiment factors: negative, neutral, and positive. The variance decomposition analysis indicates that sentiment factors account for a considerable portion of the variance in yield forecasts.

Multi-factor stock trading strategy based on DQN with multi-BiGRU and multi-head ProbSparse self-attention

Reinforcement learning is widely used in financial markets to assist investors in developing trading strategies. However, most existing models primarily focus on simple volume-price factors, and there is a need for further improvement in the returns of stock trading. To address these challenges, a multi-factor stock trading strategy based on Deep Q-Network (DQN) with Multi-layer Bidirectional Gated Recurrent Unit (Multi-BiGRU) and multi-head ProbSparse self-attention is proposed. Our strategy comprehensively characterizes the determinants of stock prices by considering various factors such as financial quality, valuation, and sentiment factors. We first use Light Gradient Boosting Machine (LightGBM) to classify turning points for stock data. Then, in the reinforcement learning strategy, Multi-BiGRU, which holds the bidirectional learning of historical data, is integrated into DQN, aiming to enhance the model’s ability to understand the dynamics of the stock market. Moreover, the multi-head ProbSparse self-attention mechanism effectively captures interactions between different factors, providing the model with deeper market insights. We validate our strategy’s effectiveness through extensive experimental research on stocks from Chinese and US markets. The results show that our method outperforms both temporal and non-temporal models in terms of stock trading returns. Ablation studies confirm the critical role of LightGBM and multi-head ProbSparse self-attention mechanism. The experiment section also demonstrates the significant advantages of our model through the presentation of box plots and statistical tests. Overall, by fully considering the multi-factor data and the model’s feature extraction capabilities, our work is expected to provide investors with more precise trading decision support.Graphical abstract

Tesla Stock Prediction Based on Twitter Comments’ Sentiment Feature Extraction and Ensemble Networks

As people become more aware of environmental issues, Tesla, a pillar of the global electric vehicle market, has become a hot commodity in recent years. The prediction of Tesla’s share price is also a hot topic in the investment market. This experiment extracts sentiment factors of tweets about Tesla’s comments, combined with Tesla’s historical stock price as the dataset for training, testing and inspection. This experiment builds time series prediction models based on LSTM, XGBoost and RF algorithms to predict Tesla’s stock price, and evaluates the prediction effectiveness of the three algorithms based on the fit and error of the prediction results. The analysis of the data shows that XGBoost has the best fit and the lowest error among the three algorithms, and that the sentiment factor has its unique utility as raw data. The experimental results also empirically demonstrate the applicability of sentiment factor analysis and the three algorithms LSTM, XGBoost and RF in the field of stock price prediction.

Media sentiment and stock returns

Based on 35,344 news articles published in the Financial Times that cover 40 companies that have been included in the Dow Jones Industrial Average, we find that a negative media sentiment in the form of a negative language tone in news articles is a priced factor in five of nine asset-pricing models that aim to explain the cross-section of stock returns. In particular, the sentiment factor is a priced factor in the market model augmented with the sentiment factor in all three samples—the 2005–09 subsample, the 2010–18 subsample, and the 2005–18 full sample—and in the Fama-French three- and five-factor models augmented with the sentiment factor in the 2010–18 subsample. In addition, factor-spanning regressions with the Fama-French five-factor model as the right-hand-side model confirm that the sentiment factor contributes to the model’s explanation of the stocks’ mean excess returns in the 2005–09 subsample and the 2005–18 full sample.

The Impact of Sentiment Scores Extracted from Product Descriptions on Customer Purchase Intention

This study investigates whether and how the textual content of product descriptions, especially the sentiment element, influences buyers’ purchase intentions. Using year-round digital transaction data from Mercari, a leading e-Commerce platform in Japan, we examine the interplay of hard and soft information signals exchanged between sellers and buyers. The study addresses two crucial questions: (1) Do the descriptions that sellers provide on product sales pages impact the buyer’s intent to purchase? and (2) In what way does the description influence the buyer’s purchase intention? Quantitative analysis is used to understand the relationship between product descriptions, sentiment elements, and purchase intentions. The results show that sentiment factors in product descriptions can serve as high-quality “signals” that can help buyers make informed purchasing decisions and reduce information asymmetry between buyers and sellers. This research contributes to understanding decision-making in online markets, particularly the role of soft information and sentiment analysis.

Is UGC sentiment helpful for recommendation? An application of sentiment-based recommendation model

PurposeGiven the critical role of user-generated content (UGC) in e-commerce, exploring various aspects of UGC can aid in understanding user purchase intention and commodity recommendation. Therefore, this study investigates the impact of UGC on purchase decisions and proposes new recommendation models based on sentiment analysis, which are verified in Douban, one of the most popular UGC websites in China.Design/methodology/approachAfter verifying the relationship between various factors and product sales, this study proposes two models, collaborative filtering recommendation model based on sentiment (SCF) and hidden factors topics recommendation model based on sentiment (SHFT), by combining traditional collaborative filtering model (CF) and hidden factors topics model (HFT) with sentiment analysis.FindingsThe results indicate that sentiment significantly influences purchase intention. Furthermore, the proposed sentiment-based recommendation models outperform traditional CF and HFT in terms of mean absolute error (MAE) and root mean square error (RMSE). Moreover, the two models yield different outcomes for various product categories, providing actionable insights for organizers to implement more precise recommendation strategies.Practical implicationsThe findings of this study advocate the incorporation of UGC sentimental factors into websites to heighten recommendation accuracy. Additionally, different recommendation strategies can be employed for different products types.Originality/valueThis study introduces a novel perspective to the recommendation algorithm field. It not only validates the impact of UGC sentiment on purchase intention but also evaluates the proposed models with real-world data. The study provides valuable insights for managerial decision-making aimed at enhancing recommendation systems.

The growth of the real estate corporate bond market in Vietnam: the role of investor sentiment

PurposeThis paper aims to examine the effects of investor sentiment on the development of the real estate corporate bond market in Vietnam.Design/methodology/approachThe research uses an autoregressive distributed lag (ARDL) model with quarterly data. Additionally, the study employs Google Trends search data (GVSI) related to topics such as “Real Estate” and “Corporate Bond” to construct a sentiment index.FindingsThe empirical outcomes reveal that real estate market sentiment improves the growth of the real estate corporate bond market, while stock market sentiment reduces it. Also, there is evidence of a long-run negative effect of corporate bond market sentiment on the total value of real estate bond issuance. Further empirical research evidences the short-term effect of sentiment and economic factors on corporate bond development in the real estate industry.Research limitations/implicationsDue to difficulties in collecting data, this paper has the limited sample of 54 valid quarterly observations. Moreover, the sentiment index based on Google search volume data only reflects the interest level of investors, not their attitudes.Practical implicationsThese results yield important implications for policymakers in respect of strengthening the corporate bond market platform and maintaining stability in macroeconomic and monetary policies in order to promote efficient and sustainable market development.Social implicationsThe study offers some suggestions for regulators and governments to improve the real estate corporate bond market.Originality/valueThis is the first quantitative study to examine the effect of sentiment factors on real estate corporate bond development in Vietnam.