Stock Recommendations Research Articles

Federated Meta Embedding Concept Stock Recommendation

Mining relevant stocks given a trending topic/concept in capital markets is an application with significant economic and societal impacts. Previous concept stock recommendation system mines concept stocks only from public social media like financial news. On stock forums, investors discuss emerging concepts and stocks by using forum comments, which are unneglectable resources to capture trending concept stocks accurately and timely. However, the comment data from a single forum is insufficient to build a high-quality recommendation system. The forums are data silos protected by privacy regulations, and their comments are still underutilized. In this paper, we propose a federated concept stock recommendation baseline and an optimized method that both leverage the private forum comments and public social media without compromising privacy regulations. Our baseline, <italic xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">i.e.</i> , Federated Meta Embedding (FedME), is built upon the federated learning framework and learns a concept-stock embedding jointly from private and public data. Our optimized method, Federated Graph Meta Embedding (FedGME), improves FedME by using a graph to combine two sources of embeddings and additional human experts' concept-stock knowledge. Empirically, the experiments on two concept stock datasets show that FedME and FedGME substantially improve the performance of recommendation. Our methods provide practical guidance on privacy-preserving FinTech applications.

Learning A-Share Stock Recommendation from Stock Graph and Historical Price Simultaneously

The Chinese stock market, marked by rapid growth and significant volatility, presents unique challenges for investors and analysts. A-share stocks, traded on the Shanghai and Shenzhen exchanges, are crucial to China’s financial system and offer opportunities for both domestic and international investors. Accurate stock recommendation tools are vital for informed decision making, especially given the ongoing regulatory changes and economic reforms in China. Current stock recommendation methods often fall short, as they typically fail to capture the complex inter-company relationships and rely heavily on financial reports, neglecting the potential of unlabeled data and historical price trends. In response, we propose a novel approach that combines graph-based structures with historical price data to develop self-learned stock embeddings for A-share recommendations. Our method leverages self-supervised learning, bypassing the need for human-generated labels and autonomously uncovering latent relationships and patterns within the data. This dual-input strategy enhances the understanding of market dynamics, leading to more accurate stock predictions. Our contributions include a novel framework for label-free stock recommendations with modeling stock connections and pricing information, and empirical evidence demonstrating the robustness and adaptability of our approach in the volatile Chinese stock market.

How do analyst recommendations on banks respond to monetary policy news? An application to the Eurozone

ABSTRACT We assess the extent to which stock analysts consider European Central Bank monetary policy decisions when formulating their investment recommendations on European banks from 2002 to 2020. To identify monetary policy shocks, we extract monetary policy factors from high-frequency rate movements surrounding monetary policy announcements. Our findings indicate that stock analysts are more concerned with non-standard (or unconventional) monetary policy measures than standard policy measures. We also demonstrate that analysts interpret restrictive (accommodative) monetary surprises as positive (negative) news for bank financial health, particularly during periods characterized by high levels of financial uncertainty. Finally, we emphasize that the reaction of stock recommendations to rate- and forward guidance-related decisions is primarily influenced by the implicit disclosure of information on the economic outlook by the ECB along with its monetary policy decision.

Sell-Side Analysts’ Assessment of ESG Risk

We assess whether and how financial analysts incorporate information about downside ESG risks. Using a unique dataset on firm-day level negative ESG incidents, we find that analysts’ outputs (i.e., stock recommendations, EPS forecasts, and target prices) are associated with negative future ESG risk events, especially those that are financially material. Further investigation suggests that analysts incorporate ESG risks not only through adjusting future cash flow expectations (i.e. the “numerator”), but also through adjusting discount rates (i.e., the “denominator”). Overall, our results highlight the ability of financial analysts to synthesize and integrate ESG risks into their research.

Government affiliation, analyst behavior and the economic consequences

AbstractA significant number of regional brokerage firms in China are owned by their respective local governments. We use this context to examine how local government affiliation affects brokerage firms’ research activities. Using a sample of analyst reports from 2013 to 2021, we find that analysts working at local government‐affiliated brokerage firms are more likely to cover firms within the local government's jurisdiction than non‐affiliated analysts. Local government‐affiliated analysts enjoy an information advantage, as evidenced by the issuance of more accurate earnings forecasts for local firms, compared to those of non‐affiliated analysts. The effect is stronger for local firms with lower information quality. Additionally, local government‐affiliated analysts show more optimism in stock recommendations, but not in earnings forecasts, for local firms, compared to non‐affiliated analysts. The effect is more pronounced for local firms that receive more local government subsidies, operate in regions with greater local government intervention or during the COVID‐19 pandemic and the National Congress conferences, and require more capital market support. The results of this study advance our understanding of the key determinants of analyst performance and provide new evidence in the debate on the role of government intervention in financial institutions.

A Systematic Review on Graph Neural Network-based Methods for Stock Market Forecasting

Financial technology (FinTech) is a field that uses artificial intelligence to automate financial services. One area of FinTech is stock analysis, which aims to predict future stock prices to develop investment strategies that maximize profits. Traditional methods of stock market prediction, such as time series analysis and machine learning, struggle to handle the non-linear, chaotic, and sudden changes in stock data and may not consider the interdependence between stocks. Recently, graph neural networks (GNNs) have been used in stock market forecasting to improve prediction accuracy by incorporating the interconnectedness of the market. GNNs can process non-Euclidean data in the form of a knowledge graph. However, financial knowledge graphs can have dynamic and complex interactions, which can be challenging for graph modeling technologies. This work presents a systematic review of graph-based approaches for stock market forecasting. This review covers different types of stock analysis tasks (classification, regression, and stock recommendation), a generalized framework for solving these tasks, and a review of various features, datasets, graph models, and evaluation metrics used in the stock market. The results of various studies are analyzed, and future directions for research are highlighted.

Research on information fusion of security analysts’ stock recommendations based on two-dimensional D-S evidence theory

Security analysts play a vital role as an information intermediary in the stock market. Their stock recommendations are important references for investors. The efficiency of investment decision-making could be improved by judging the reliability of stock recommendations based on analyst characteristics and fusing the recommendations. We propose an information fusion method for security analysts’ stock recommendations based on two-dimensional Dempster-Shafer (D-S) evidence theory, which comprehensively considers the external and internal characteristics of analysts. The characteristics of analysts are used to measure the reliability of the stock recommendations and modify the evidence, then the D-S fusion rule is used for evidence fusion. Compared with the forecast results of statistical methods and machine learning methods, the two-dimensional D-S evidence theory model we proposed has a higher forecast accuracy, which effectively improves the information efficiency of the stock market and helps investors to make decisions efficiently and scientifically.

ESG performance and analyst recommendations: evidence from sustainability analysts in the Chinese market

PurposeIn this study, we examine the influence of a firm’s environmental, social, and governance (ESG) performance on analysts’ stock recommendations and earnings forecast accuracy in the Chinese context.Design/methodology/approachWe take a textual analysis approach to analyst research reports issued between 2010 and 2019, and differentiate between two distinct analyst categories: “sustainability analysts,” which refer to those more inclined to incorporate ESG information into their analyses, and “other analysts.”FindingsOur evidence indicates that sustainability analysts tend to be significantly more likely than others to provide positive stock recommendations and demonstrate enhanced accuracy in forecasting earnings for companies with superior ESG performance. Our additional analyses reveal that this finding is particularly prominent for analysts who graduated from institutions emphasizing the protection of the environment, those recognized as star analysts, those affiliated with ESG-oriented brokerages, and forecasts made by analysts in the later part of the sample period. Our findings further indicate that sustainability analysts exhibit a more pronounced negative response when confronted with a negative ESG event.Originality/valueIn general, the evidence from this study reveals the interplay between ESG factors and analyst behavior, offering valuable implications for both financial analysts and sustainable investment strategies.

Aggregate uncertainty, information acquisition, and analyst stock recommendations

AbstractWe examine the informativeness of analyst stock recommendations in the presence of aggregate uncertainty. Our results suggest that a one standard deviation increase in aggregate uncertainty decreases the likelihood of influential recommendation revisions by 5.26%. Increased aggregate uncertainty leads to a small stock price impact for upgrade and downgrade recommendations. Our findings reveal consistent search for information by investors, which, support a post‐recommendation price drift amidst high aggregate uncertainty. We further find that investors of firms with fewer distracted shareholders, less readable financial statements, and more informed trading seek more information when aggregate uncertainty is high. Our study highlights that investors become more cautious while responding to analysts' stock recommendations during high aggregate uncertainty.

Does analyst ESG experience matter?

We examine the relationship between analysts' task-specific experience in the context of ESG information and the informativeness of their stock recommendation revisions. While sell-side analysts incorporate ESG information in their valuation process and research reports, previous studies have indicated that the increased availability of ESG information in the market poses challenges for analysts to issue incrementally informative reports. Building upon existing literature that highlights systematic differences in analysts' performance that is attributed to their experience, we introduce a measure of financial analysts' ESG experience. We document a positive association between our proxy of analyst ESG experience and market reactions to their recommendation revisions. Our findings also show that analysts' ESG experience contributes to the interpretation of information included in firm ESG reports. We further find support for a stronger association between the market reaction and our ESG-experience proxy when firms exhibit lower levels of ESG disclosure and when they face higher external scrutiny due to their exposure to ESG-related risks. Our results are robust when considering analysts' innate and forecasting ability and when additional specifications are applied.

Integrated Reporting and the Informativeness of Financial Analysts’ Stock Recommendations

Synopsis The research problem The purpose of this study was to examine the market reaction to sell-side analyst recommendation revisions issued under an integrated reporting (IR) approach. Motivation Advocates of this corporate reporting approach argue that IR enhances capital markets’ information environment by rendering investors better able to assess the value creation process of a firm. Recent empirical studies corroborate this argument. Considering the central role of financial intermediaries, we investigated whether the informativeness of analyst recommendation revisions is associated with the adoption of IR and the quality of integrated reports. The test hypotheses We tested whether the informativeness of analyst recommendation revisions decreases or increases after the mandatory adoption of an IR approach. Furthermore, we tested whether the informativeness is negatively or positively related to the quality of the released integrated report after the mandatory adoption of an IR approach. Target population We focused on the South African capital market, which is the only setting where IR is mandated. We utilized a sample of 3,201 recommendation revisions made within a 3-year window around the mandatory IR adoption. Adopted methodology This study used ordinary least square (OLS) regressions and applied difference-in-differences as well as instrumental variable approaches. Analyses We modeled the market reaction to the recommendation revisions as a function of the period in which the recommendations are announced (i.e., pre- or post-adoption), along with other factors affecting market reaction. In subsequent tests, we also modeled the market reaction to the recommendation revisions as a function of the quality of a firm’s integrated report. Findings We found strong evidence that analysts’ revisions exhibited economically and statistically significantly lower information content under IR. Moreover, we found that upgrades and downgrades issued in the post-adoption period were less informative when issued for firms with high-quality integrated reports. Overall, results showed that the benefit of acquiring advice from analysts became more marginal under an IR approach.

Politically Affiliated Analysts

Government ownership of financial intermediaries is pervasive around the world. In this study, we examine the impact of common government ownership between the brokerage and listed firms on the information production role of brokerage firms. We show that affiliated analysts tend to issue more optimistic recommendations for stocks of firms controlled by the same government entity that controls their brokerage firms. This optimistic bias is particularly pronounced during periods of economic shocks. Our study demonstrates this by utilizing additional tariff impositions and tariff exemptions during the U.S.–China trade war as exogenous negative and positive shocks, respectively. Additionally, our study indicates that stocks recommended by politically affiliated analysts tend to underperform those recommended by independent analysts, implying that the optimism stems from conflicts of interest rather than superior information. Furthermore, our research highlights that sophisticated investors perceive the potential bias and incorporate it into their trading. Consistent with an exchange of favors story, politically affiliated brokerage firms receive a larger allocation during the issuance of local government debt, whereas governments subscribe for more shares during seasoned equity offerings by these affiliated brokerage firms. This paper was accepted by Kay Giesecke, finance. Funding: The authors gratefully acknowledge the financial support from the National Natural Science Foundation of China [Grants 71972088, 72132002, and 71991473], the National Social Science Foundation of China [Grants 21ZDA010 and 22VRC145], and the Innovation and Talent Base for Digital Technology and Finance [Grant B21038]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.00579 .

Integrated Financial Analysis and Asset Recommendation

Integrated Financial Analysis and Asset Recommendation

Can Perturbations Help Reduce Investment Risks? Risk-aware Stock Recommendation via Split Variational Adversarial Training

In the stock market, a successful investment requires a good balance between profits and risks. Based on the learning to rank paradigm, stock recommendation has been widely studied in quantitative finance to recommend stocks with higher return ratios for investors. Despite the efforts to make profits, many existing recommendation approaches still have some limitations in risk control, which may lead to intolerable paper losses in practical stock investing. To effectively reduce risks, we draw inspiration from adversarial learning and propose a novel Split Variational Adversarial Training (SVAT) method for risk-aware stock recommendation. Essentially, SVAT encourages the stock model to be sensitive to adversarial perturbations of risky stock examples and enhances the model’s risk awareness by learning from perturbations. To generate representative adversarial examples as risk indicators, we devise a variational perturbation generator to model diverse risk factors. Particularly, the variational architecture enables our method to provide a rough risk quantification for investors, showing an additional advantage of interpretability. Experiments on several real-world stock market datasets demonstrate the superiority of our SVAT method. By lowering the volatility of the stock-recommendation model, SVAT effectively reduces investment risks and outperforms state-of-the-art baselines by more than 30% in terms of risk-adjusted profits. All the experimental data and source code are available at https://drive.google.com/drive/folders/14AdM7WENEvIp5x5bV3zV_i4Aev21C9g6?usp=sharing .

Gender Differences in Analyst Stock Recommendations and Market Reactions

This paper provides evidence of gender heterogeneity in stock recommendations issued by analysts in China. The study finds that compared to male analysts, female analysts tend to have more conservative stock recommendations and lower levels of optimism. Male analysts are more likely to issue "buy" and "hold" recommendations, while female analysts are more likely to issue more conservative "neutral," "sell," and "reduce" recommendations. In the short term, the market reacts strongly to the ratings of female analysts, but their optimistic recommendations receive a poor response from investors. Investors may hold stereotypical beliefs towards female analysts, expecting them to adopt conservative strategies, and thus underestimate them when they make relatively optimistic suggestions. However, in the long term, there is no evident bias against female analysts in the market.

Incorporating Domain-Specific Traits into Personality-Aware Recommendations for Financial Applications

The general personality traits, notably the Big-Five personality traits, have been increasingly integrated into recommendation systems. The personality-aware recommendations, which incorporate human personality into recommendation systems, have shown promising results in general recommendation areas including music, movie, and e-commerce recommendations. On the other hand, the number of research delving into the applicability of personality-aware recommendations in specialized domains such as finance and education remains limited. In addition, these domains have unique challenges in incorporating personality-aware recommendations as domain-specific psychological traits such as risk tolerance and behavioral biases play a crucial role in explaining user behavior in these domains. Addressing these challenges, this study addresses an in-depth exploration of personality-aware recommendations in the financial domain, specifically within the context of stock recommendations. First, this study investigates the benefits of deploying general personality traits in stock recommendations through the integration of personality-aware recommendations with user-based collaborative filtering approaches. Second, this study further verifies whether incorporating domain-specific psychological traits along with general personality traits enhances the performance of stock recommender systems. Thirdly, this paper introduces a personalized stock recommendation model that incorporates both general personality traits and domain-specific psychological traits as well as transaction data. The experimental results show that the proposed model outperformed baseline models in financial stock recommendations.

FollowAKOInvestor: Stock recommendation by hearing voices from all kinds of investors with machine learning

As an increasing number of investors share their opinions on social networks, a critical challenge is to provide advice and assistance in making well-informed investment decisions by considering enormous online sentiments. A typical way is to aggregate investors’ sentiments into an investing signal. Most previous studies of sentiment aggregation rely on human-defined rules summarized from experience, which are widely accepted for simplicity and interpretability. However, the relations between investor sentiments and wise investment decisions generated by integrating sentiments could be complex and counter-intuitive, which may not be fully described by human-defined rules and require further investigation. This paper proposes a novel machine learning-based method called FollowAKOInvestor to integrate investors’ sentiments more effectively for highly profitable stock recommendations. It divides investors into groups by clustering algorithm or their expertise levels, then extracts sentiment features from each group to train a machine learning model. The outputs are utilized to score stocks and provide investment recommendations. Based on an extensive dataset spanning from January 2017 to December 2019, sourced from StockTwits, we simulate a one-year recommendation procedure to compare and evaluate the performance of the proposed method and baselines. The experimental results support the superiority of machine learning over human-defined rules in making the most of sentiments from various investors for stock recommendations.

Stock recommendation methods for stability

Stock recommendation methods for stability

A Static-Dynamic Hypergraph Neural Network Framework Based on Residual Learning for Stock Recommendation

Stock ranking prediction is an effective method for achieving a high investment return and plays a crucial role in investment decisions. However, previous studies have overlooked the interconnections among stocks or have solely relied on predefined graphs for stock relationship information. The predefined graphs may not capture all possible relationships and may not be suitable for describing dynamic relationships. To address these issues, we propose a Static-Dynamic hypergraph neural network framework based on Residual Learning (SD-RL). Compared with traditional methods, SD-RL has the following advantages. (1) Stocks are no longer treated as isolated entities; instead, their static and dynamic relationship information is taken into account. (2) Leveraging the data-driven methodology, SD-RL autonomously learns both the static graph and dynamic hypergraph through dedicated graph learning and hypergraph learning modules, respectively. (3) By employing residual learning, various latent relationship information flows are mined, which enhances the stock embedding’s capacity to capture trends. Extensive experiments on the real data verify the effectiveness of our proposed methods.

Chief executive officer narcissism, corporate inertia, and securities analysts’ stock recommendations

The narcissism of chief executive officers is attracting much research interest because of its potential effects on the strategic decisions, financial performance, and competitive standing of firms. This article addresses a significant gap in the literature by analyzing the effect of chief executive officer narcissism on security analysts’ stock recommendations. As financial-market intermediaries between firms and investors, analysts are an important corporate-governance actor, whose stock recommendations are consequential for the market value of a firm. Drawing on the idea of observers’ implicit leadership theories, we argue that greater chief executive officer narcissism will predict lower analysts’ stock recommendations because narcissistic chief executive officers’ penchant for risk-taking will lead analysts to categorize them as ineffective leaders. We argue further that signals of corporate inertia conveyed by the age, size, and reputation of firms will positively moderate the chief executive officer narcissism–analysts’ stock recommendation relationship, because analysts will expect inertia to be offset by narcissistic chief executive officers’ risk-taking, a dynamic likely to improve firm performance. US panel data provide support for the theorized chief executive officer narcissism–analysts’ stock recommendations relationship and indicates significant moderation effects of the reputation and size of firms. The article discusses the study’s contributions and implications for research and practice.