|
on Big Data |
| By: | Mateo Seré |
| Abstract: | This study, utilizing a novel dataset from economic seminar audio recordings, investigates gender-based peer interactions, structured around five key findings: (i) Female speakers are interrupted more frequently, earlier, and differently than males; (ii) the extra interruptions largely stem from female, not male, audience members; (iii) male participants pose fewer questions but more comments to female presenters; (iv) audience members of both genders interrupt female speakers with a more negative tone; (v) less senior female presenters receive more interruptions from women. Control variables include seminar series, presentation topic, and factors like presenter affiliation, seniority, and department ranking. |
| Date: | 2023–07 |
| URL: | http://d.repec.org/n?u=RePEc:hdl:wpaper:2309&r=big |
| By: | Jean Lee; Nicholas Stevens; Soyeon Caren Han; Minseok Song |
| Abstract: | Large Language Models (LLMs) have shown remarkable capabilities across a wide variety of Natural Language Processing (NLP) tasks and have attracted attention from multiple domains, including financial services. Despite the extensive research into general-domain LLMs, and their immense potential in finance, Financial LLM (FinLLM) research remains limited. This survey provides a comprehensive overview of FinLLMs, including their history, techniques, performance, and opportunities and challenges. Firstly, we present a chronological overview of general-domain Pre-trained Language Models (PLMs) through to current FinLLMs, including the GPT-series, selected open-source LLMs, and financial LMs. Secondly, we compare five techniques used across financial PLMs and FinLLMs, including training methods, training data, and fine-tuning methods. Thirdly, we summarize the performance evaluations of six benchmark tasks and datasets. In addition, we provide eight advanced financial NLP tasks and datasets for developing more sophisticated FinLLMs. Finally, we discuss the opportunities and the challenges facing FinLLMs, such as hallucination, privacy, and efficiency. To support AI research in finance, we compile a collection of accessible datasets and evaluation benchmarks on GitHub. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.02315&r=big |
| By: | Yulu Gong; Mengran Zhu; Shuning Huo; Yafei Xiang; Hanyi Yu |
| Abstract: | In the age of the Internet, people's lives are increasingly dependent on today's network technology. However, network technology is a double-edged sword, bringing convenience to people but also posing many security challenges. Maintaining network security and protecting the legitimate interests of users is at the heart of network construction. Threat detection is an important part of a complete and effective defense system. In the field of network information security, the technical update of network attack and network protection is spiraling. How to effectively detect unknown threats is one of the concerns of network protection. Currently, network threat detection is usually based on rules and traditional machine learning methods, which create artificial rules or extract common spatiotemporal features, which cannot be applied to large-scale data applications, and the emergence of unknown threats causes the detection accuracy of the original model to decline. With this in mind, this paper uses deep learning for advanced threat detection to improve cybersecurity resilienc e in the financial industry. Many network security researchers have shifted their focus to exceptio n-based intrusion detection techniques. The detection technology mainly uses statistical machine learning methods - collecting normal program and network behavior data, extracting multidimensional features, and training decision machine learning models on this basis (commonly used include naive Bayes, decision trees, support vector machines, random forests, etc.). In the detection phase, program code or network behavior that deviates from the normal value beyond the tolerance is considered malicious code or network attack behavior. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.09820&r=big |
| By: | Mestiri, Sami |
| Abstract: | In the last years, the financial sector has seen an increase in the use of machine learning models in banking and insurance contexts. Advanced analytic teams in the financial community are implementing these models regularly. In this paper, we analyses the limitations of machine learning methods, and then provides some suggestions on the choice of methods in financial applications. We refer the reader to the R libraries that can be used to compute the Machine learning methods |
| Keywords: | Financial applications; Machine learning ; R software. |
| JEL: | C45 C5 G23 |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:119998&r=big |
| By: | Sven Klaassen; Jan Teichert-Kluge; Philipp Bach; Victor Chernozhukov; Martin Spindler; Suhas Vijaykumar |
| Abstract: | This paper explores the use of unstructured, multimodal data, namely text and images, in causal inference and treatment effect estimation. We propose a neural network architecture that is adapted to the double machine learning (DML) framework, specifically the partially linear model. An additional contribution of our paper is a new method to generate a semi-synthetic dataset which can be used to evaluate the performance of causal effect estimation in the presence of text and images as confounders. The proposed methods and architectures are evaluated on the semi-synthetic dataset and compared to standard approaches, highlighting the potential benefit of using text and images directly in causal studies. Our findings have implications for researchers and practitioners in economics, marketing, finance, medicine and data science in general who are interested in estimating causal quantities using non-traditional data. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.01785&r=big |
| By: | J. van den Berg, Gerard (IFAU and University of Gronigen); Kunaschk, Max (IAB Nuremberg); Lang, Julia (IAB Nuremberg); Stephan, Gesine (IAB Nuremberg); Uhlendorff, Arne (CNRS and CREST i Paris) |
| Abstract: | We analyze unique data on three sources of information on the probability of re-employment within 6 months (RE6), for the same individuals sampled from the inflow into unemployment. First, they were asked for their perceived probability of RE6. Second, their caseworkers revealed whether they expected RE6. Third, random-forest machine learning methods are trained on admin istrative data on the full inflow, to predict individual RE6. We compare the predictive performance of these measures and consider how combinations im prove this performance. We show that self-reported (and to a lesser extent caseworker) assessments sometimes contain information not captured by the machine learning algorithm. |
| Keywords: | Unemployment; expectations; prediction; random forest; unemloyment insurance; information; |
| JEL: | C21 C41 C53 C55 J64 J65 |
| Date: | 2023–11–10 |
| URL: | http://d.repec.org/n?u=RePEc:hhs:ifauwp:2023_022&r=big |
| By: | Magassouba, Aboubacar Sidiki; Diallo, Abdourahmane; Nkurunziza, Armel; Tchole, Ali Issakou Malam; Touré, Almamy Amara; Magassouba, Mamoudou; Sylla, Younoussa; diallo, Mamadou Abdoulaye R; Nabé, Aly Badara |
| Abstract: | Background: Contraceptive continuation is crucial for assessing the quality and effectiveness of family planning programs, yet it remains challenging, particularly in sub-Saharan Africa. Traditional statistical methods may only partially capture complex relationships and interactions among variables. Machine learning, an artificial intelligence domain, offers the potential to handle large and intricate datasets, uncover hidden patterns, make accurate predictions, and provide interpretable results. Objective: Using data from the last four Demographic and Health Surveys, our study utilised a machine learning model to predict contraceptive continuation among women aged 15-49 in a West African country. Additionally, we employed SHAP (SHapley Additive exPlanations) analysis to identify and rank the most influential features for the prediction. Methods: We employed LightGBM, a gradient-boosting framework that employs tree-based learning algorithms, to construct our predictive model. Our multilevel LGBM model accounted for country-level variations while controlling for individual variables. Furthermore, optimization techniques were utilized to enhance performance and computation efficiency. Hyperparameter tuning was conducted using Optuna, and the machine learning model performance was evaluated based on accuracy and area under the curve (AUC) metrics. SHAP analysis was employed to elucidate the model's predictions and feature impacts. Results: Our final model demonstrated an accuracy of 74% and an AUC of 82%, highlighting its effectiveness in predicting contraceptive continuation among women aged 15-49. The most influential features for prediction encompassed the number of children under 5 in household, age, desire for more children, current family planning method type, total children ever born, household relationship structure, recent health facility visits, country, and husband's desire for children. Conclusion: Machine learning is a valuable tool for accurately predicting and interpreting contraceptive continuation among women in sub-Saharan Africa. The identified influential features offer insights for designing interventions tailored to different groups, catering to their specific needs and preferences, and ultimately improving reproductive health outcomes.Keywords Contraceptive continuation, Machine learning, Sub-Saharan Africa, Predictive model |
| Date: | 2024–01–26 |
| URL: | http://d.repec.org/n?u=RePEc:osf:socarx:u38sh&r=big |
| By: | Niko Hauzenberger; Massimiliano Marcellino; Michael Pfarrhofer; Anna Stelzer |
| Abstract: | We propose and discuss Bayesian machine learning methods for mixed data sampling (MIDAS) regressions. This involves handling frequency mismatches with restricted and unrestricted MIDAS variants and specifying functional relationships between many predictors and the dependent variable. We use Gaussian processes (GP) and Bayesian additive regression trees (BART) as flexible extensions to linear penalized estimation. In a nowcasting and forecasting exercise we focus on quarterly US output growth and inflation in the GDP deflator. The new models leverage macroeconomic Big Data in a computationally efficient way and offer gains in predictive accuracy along several dimensions. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.10574&r=big |
| By: | Teodoro Baldazzi (Università Roma Tre); Luigi Bellomarini (Bank of Italy); Stefano Ceri (Politecnico di Milano); Andrea Colombo (Politecnico di Milano); Andrea Gentili (Bank of Italy); Emanuel Sallinger (TU Wien; University of Oxford) |
| Abstract: | Large Language Models (LLMs) usually undergo a pre-training process on extensive collections of generic textual data, which are often publicly accessible. Pre-training enables LLMs to grasp language grammar, understand context, and convey a sense of common knowledge. Pre-training can be likened to machine learning training: the LLM is trained to predict the next basic text unit (e.g., a word or a sequence of words) based on the sequence of previously observed units. However, despite the impressive generalization and human-like interaction capabilities shown in Natural Language Processing (NLP) tasks, pre-trained LLMs exhibit significant limitations and provide poor accuracy when applied in specialized domains. Their main limitation stems from the fact that data used in generic pre-training often lacks knowledge related to the specific domain. To address these limitations, fine-tuning techniques are often employed to refine pre-trained models using domain-specific data. Factual information is extracted from company databases to create text collections for fine-tuning purposes. However, even in this case, results tend to be unsatisfactory in complex domains, such as financial markets and finance in general. Examining the issue from a different perspective, the Knowledge Representation and Reasoning (KRR) community has focused on producing formalisms, methods, and systems for representing complex Enterprise Knowledge. In particular, Enterprise Knowledge Graphs (EKGs) can leverage a combination of factual information in databases and business knowledge specified in a compact and formal fashion. EKGs serve the purpose of answering specific domain queries through established techniques such as ontological reasoning. Domain knowledge is represented in symbolic forms, e.g., logic-based languages, and used to draw consequential conclusions from the available data. However, while EKGs are applied successfully in many financial scenarios, they lack flexibility, common sense and linguistic orientation, essential for NLP. This paper proposes an approach aimed at enhancing the utility of LLMs for specific applications, such as those related to financial markets. The approach involves guiding the fine-tuning process of LLMs through ontological reasoning on EKGs. In particular, we exploit the Vadalog system and its language, a state-of-the-art automated reasoning framework, to synthesize an extensive fine- tuning corpus from a logical formalization of domain knowledge in an EKG. Our contribution consists of a technique called verbalization, which transforms the set of inferences determined by ontological reasoning into a corpus for fine-tuning. We present a complete software architecture that applies verbalization to four NLP tasks: question answering, i.e., providing accurate responses in a specific domain in good prose; explanation, i.e., systematically justifying the conclusions drawn; translation, i.e., converting domain specifications into logical formalization; and description, i.e., explaining formal specifications in prose. We apply the approach and our architecture in the context of financial markets, presenting a proof of concept that highlights their advantages. |
| Keywords: | Ontological reasoning, Large language models, Knowledge graphs |
| Date: | 2024–01 |
| URL: | http://d.repec.org/n?u=RePEc:bdi:wpmisp:mip_044_24&r=big |
| By: | Takuji Arai; Yuto Imai |
| Abstract: | This paper aims to develop a supervised deep-learning scheme to compute call option prices for the Barndorff-Nielsen and Shephard model with a non-martingale asset price process having infinite active jumps. In our deep learning scheme, teaching data is generated through the Monte Carlo method developed by Arai and Imai (2024). Moreover, the BNS model includes many variables, which makes the deep learning accuracy worse. Therefore, we will create another input variable using the Black-Scholes formula. As a result, the accuracy is improved dramatically. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.00445&r=big |
| By: | Phoebe Koundouri; Panagiotis Stavros Aslanidis; Konstantinos Dellis; Georgios Feretzakis; Angelos Plataniotis |
| Abstract: | This paper introduces a machine learning (ML) based approach for integrating Human Security (HS) and Sustainable Development Goals (SDGs). Originating in the 1990s, HS focuses on strategic, people-centric interventions for ensuring comprehensive welfare and resilience. It closely aligns with the SDGs, together forming the foundation for global sustainable development initiatives. Our methodology involves mapping 44 reports to the 17 SDGs using expert-annotated keywords and advanced ML techniques, resulting in a web-based SDG mapping tool. This tool is specifically tailored for the HS-SDG nexus, enabling the analysis of 13 new reports and their connections to the SDGs. Through this, we uncover detailed insights and establish strong links between the reports and global objectives, offering a nuanced understanding of the interplay between HS and sustainable development. This research provides a scalable framework to explore the relationship between HS and the Paris Agenda, offering a practical, efficient resource for scholars and policymakers. |
| Keywords: | Artificial Intelligence in Policy Making, Data Mining, Human-Centric Governance Strategies, Human Security, Machine Learning, Sustainable Development Goals |
| Date: | 2024–02–20 |
| URL: | http://d.repec.org/n?u=RePEc:aue:wpaper:2406&r=big |
| By: | Kelvin J. L. Koa; Yunshan Ma; Ritchie Ng; Tat-Seng Chua |
| Abstract: | Explaining stock predictions is generally a difficult task for traditional non-generative deep learning models, where explanations are limited to visualizing the attention weights on important texts. Today, Large Language Models (LLMs) present a solution to this problem, given their known capabilities to generate human-readable explanations for their decision-making process. However, the task of stock prediction remains challenging for LLMs, as it requires the ability to weigh the varying impacts of chaotic social texts on stock prices. The problem gets progressively harder with the introduction of the explanation component, which requires LLMs to explain verbally why certain factors are more important than the others. On the other hand, to fine-tune LLMs for such a task, one would need expert-annotated samples of explanation for every stock movement in the training set, which is expensive and impractical to scale. To tackle these issues, we propose our Summarize-Explain-Predict (SEP) framework, which utilizes a self-reflective agent and Proximal Policy Optimization (PPO) to let a LLM teach itself how to generate explainable stock predictions in a fully autonomous manner. The reflective agent learns how to explain past stock movements through self-reasoning, while the PPO trainer trains the model to generate the most likely explanations from input texts. The training samples for the PPO trainer are also the responses generated during the reflective process, which eliminates the need for human annotators. Using our SEP framework, we fine-tune a LLM that can outperform both traditional deep-learning and LLM methods in prediction accuracy and Matthews correlation coefficient for the stock classification task. To justify the generalization capability of our framework, we further test it on the portfolio construction task, and demonstrate its effectiveness through various portfolio metrics. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.03659&r=big |
| By: | Ozili, Peterson K |
| Abstract: | Artificial intelligence (AI) is a topic of interest in the finance literature. However, its role and implications for central banks have not received much attention in the literature. Using discourse analysis method, this article identifies the benefits and risks of artificial intelligence in central banking. The benefits of artificial intelligence for central banks are that deploying artificial intelligence systems will encourage central banks to develop information technology (IT) and data science capabilities, it will assist central banks in detecting financial stability risks, it will aid the search for granular micro economic/non-economic data from the internet so that the data can support central banks in making policy decisions, it enables the use of AI-generated synthetic data, and it enables task automation in central banking operations. However, the use of artificial intelligence in central banking poses some risks which include data privacy risk, the risk that using synthetic data could lead to false positives, high risk of embedded bias, difficulty of central banks to explain AI-based policy decisions, and cybersecurity risk. The article also offers some considerations for responsible use of artificial intelligence in central banking. |
| Keywords: | central bank, artificial intelligence, financial stability, responsible AI, artificial intelligence model. |
| JEL: | E51 E52 E58 |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:120151&r=big |
| By: | Beck, Günter W.; Carstensen, Kai; Menz, Jan-Oliver; Schnorrenberger, Richard; Wieland, Elisabeth |
| Abstract: | We study how millions of highly granular and weekly household scanner data combined with novel machine learning techniques can help to improve the nowcast of monthly German inflation in real time. Our nowcasting exercise targets three hierarchy levels of the official consumer price index. First, we construct a large set of weekly scanner-based price indices at the lowest aggregation level underlying official German inflation, such as those of butter and coffee beans. We show that these indices track their official counterparts extremely well. Within a mixed-frequency modeling framework, we also demonstrate that these scanner-based price indices improve inflation nowcasts at this very narrow level, notably already after the first seven days of a month. Second, we apply shrinkage estimators to exploit the large set of scanner-based price indices in nowcasting product groups such as processed and unprocessed food. This yields substantial predictive gains compared to a time series benchmark model. Finally, we nowcast headline inflation. Adding high-frequency information on energy and travel services, we construct highly competitive nowcasting models that are on par with, or even outperform, survey-based inflation expectations that are notoriously difficult to beat. |
| Keywords: | Inflationnowcasting, machine learningmethods, scannerprice data, mixed-frequency modeling |
| JEL: | E31 C55 E37 C53 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:zbw:bubdps:282982&r=big |
| By: | Michael Levere; David Wittenburg |
| Abstract: | Children’s participation in the federal Supplemental Security Income (SSI) program has declined substantially over the past decade. Many children with disabilities might be eligible for SSI, yet barriers such as a lack of knowledge of the program or perceived challenges with applying may limit participation. In this paper, we use machine learning models on Medicaid administrative data to estimate the number and characteristics of children who are potentially eligible for SSI but do not currently receive benefits. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:crr:crrwps:wp2023-20&r=big |
| By: | Marcos Lacasa Cazcarra |
| Abstract: | This paper analyzes the impact of the National Minimum Wage from 2001 to 2021. The MNW increased from 505.7/month (2001) to 1, 108.3/month (2021). Using the data provided by the Spanish Tax Administration Agency, databases that represent the entire population studied can be analyzed. More accurate results and more efficient predictive models are provided by these counts. This work is characterized by the database used, which is a national census and not a sample or projection. Therefore, the study reflects results and analyses based on historical data from the Spanish Salary Census 2001-2021. Various machine-learning models show that income inequality has been reduced by raising the minimum wage. Raising the minimum wage has not led to inflation or increased unemployment. On the contrary, it has been consistent with increased net employment, contained prices, and increased corporate profit margins. The most important conclusion is that an increase in the minimum wage in the period analyzed has led to an increase in the wealth of the country, increasing employment and company profits, and is postulated, under the conditions analyzed, as an effective method for the redistribution of wealth. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.02402&r=big |
| By: | Cai, Hengrui; Shi, Chengchun; Song, Rui; Lu, Wenbin |
| Abstract: | An individualized decision rule (IDR) is a decision function that assigns each individual a given treatment based on his/her observed characteristics. Most of the existing works in the literature consider settings with binary or finitely many treatment options. In this paper, we focus on the continuous treatment setting and propose a jump interval-learning to develop an individualized interval-valued decision rule (I2DR) that maximizes the expected outcome. Unlike IDRs that recommend a single treatment, the proposed I2DR yields an interval of treatment options for each individual, making it more flexible to implement in practice. To derive an optimal I2DR, our jump interval-learning method estimates the conditional mean of the outcome given the treatment and the covariates via jump penalized regression, and derives the corresponding optimal I2DR based on the estimated outcome regression function. The regressor is allowed to be either linear for clear interpretation or deep neural network to model complex treatment-covariates interactions. To implement jump interval-learning, we develop a searching algorithm based on dynamic programming that efficiently computes the outcome regression function. Statistical properties of the resulting I2DR are established when the outcome regression function is either a piecewise or continuous function over the treatment space. We further develop a procedure to infer the mean outcome under the (estimated) optimal policy. Extensive simulations and a real data application to a Warfarin study are conducted to demonstrate the empirical validity of the proposed I2DR. |
| Keywords: | continuous treatment; dynamic programming; individualized interval-valued decision rule; jump interval-learning; precision medicine |
| JEL: | C1 |
| Date: | 2023–02–13 |
| URL: | http://d.repec.org/n?u=RePEc:ehl:lserod:118231&r=big |
| By: | Daniel Borup; Philippe Goulet Coulombe; Erik Christian Montes Schütte; David E. Rapach; Sander Schwenk-Nebbe |
| Abstract: | We introduce the performance-based Shapley value (PBSV) to measure the contributions of individual predictors to the out-of-sample loss for time-series forecasting models. Our new metric allows a researcher to anatomize out-of-sample forecasting accuracy, thereby providing valuable information for interpreting time-series forecasting models. The PBSV is model agnostic—so it can be applied to any forecasting model, including "black box" models in machine learning, and it can be used for any loss function. We also develop the TS-Shapley-VI, a version of the conventional Shapley value that gauges the importance of predictors for explaining the in-sample predictions in the entire sequence of fitted models that generates the time series of out-of-sample forecasts. We then propose the model accordance score to compare predictor ranks based on the TS-Shapley-VI and PBSV, thereby linking the predictors' in-sample importance to their contributions to out-of-sample forecasting accuracy. We illustrate our metrics in an application forecasting US inflation. |
| Keywords: | model interpretation; Shapley value; predictor importance; loss function; machine learning; inflation |
| JEL: | C22 C45 C52 C53 E31 E37 |
| Date: | 2024–02–21 |
| URL: | http://d.repec.org/n?u=RePEc:fip:fedawp:97785&r=big |
| By: | Lisa Bruttel (University of Potsdam, CEPA); Gerald Eisenkopf (University of Vechta); Juri Nithammer (University of Potsdam) |
| Abstract: | Leadership plays an important role for the efficient and fair solution of social dilemmas but the effectiveness of a leader can vary substantially. Two main factors of leadership impact are the ability to induce high contributions by all group members and the (expected) fair use of power. Participants in our experiment decide about contributions to a public good. After all contributions are made, the leader can choose how much of the joint earnings to assign to herself; the remainder is distributed equally among the followers. Using machine learning techniques, we study whether the content of initial open statements by the group members predicts their behavior as a leader and whether groups are able to identify such clues and endogenously appoint a “good” leader to solve the dilemma. We find that leaders who promise fairness are more likely to behave fairly, and that followers appoint as leaders those who write more explicitly about fairness and efficiency. However, in their contribution decision, followers focus on the leader’s first-move contribution and place less importance on the content of the leader’s statements. |
| Keywords: | Leadership, Public good, Voting, Promises, Experiment |
| JEL: | C92 D23 D72 D83 |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:pot:cepadp:73&r=big |
| By: | Xiaorui Zuo; Yao-Tsung Chen; Wolfgang Karl H\"ardle |
| Abstract: | In the burgeoning realm of cryptocurrency, social media platforms like Twitter have become pivotal in influencing market trends and investor sentiments. In our study, we leverage GPT-4 and a fine-tuned transformer-based BERT model for a multimodal sentiment analysis, focusing on the impact of emoji sentiment on cryptocurrency markets. By translating emojis into quantifiable sentiment data, we correlate these insights with key market indicators like BTC Price and the VCRIX index. This approach may be fed into the development of trading strategies aimed at utilizing social media elements to identify and forecast market trends. Crucially, our findings suggest that strategies based on emoji sentiment can facilitate the avoidance of significant market downturns and contribute to the stabilization of returns. This research underscores the practical benefits of integrating advanced AI-driven analyses into financial strategies, offering a nuanced perspective on the interplay between digital communication and market dynamics in an academic context. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.10481&r=big |
| By: | Johannes Carow (Johannes Gutenberg University Mainz); Niklas M. Witzig (Johannes Gutenberg University Mainz) |
| Abstract: | We study the impact of time pressure on strategic risk-taking of professional chess players. We propose a novel machine-learning-based measure for the degree of strategic risk of a single chess move and apply this measure to the 2013-2023 FIDE Chess World Cups that allow for plausibly exogenous variation in thinking time. Our results indicate that time pressure leads chess players to opt for more risk-averse moves. We additionally provide correlational evidence for strategic loss aversion, a tendency for risky moves after a mistake/ in a disadvantageous position. This suggests that high-proficiency decision-makers in highstake situations react to time pressure and contextual factors more broadly. We discuss the origins and implication of this finding in our setting. |
| Keywords: | Chess, Risk, Time Pressure, Loss Aversion, Machine Learning |
| JEL: | C26 C45 D91 |
| Date: | 2024–02–22 |
| URL: | http://d.repec.org/n?u=RePEc:jgu:wpaper:2404&r=big |
| By: | Yike Wang; Chris Gu; Taisuke Otsu |
| Abstract: | This paper presents a novel application of graph neural networks for modeling and estimating network heterogeneity. Network heterogeneity is characterized by variations in unit's decisions or outcomes that depend not only on its own attributes but also on the conditions of its surrounding neighborhood. We delineate the convergence rate of the graph neural networks estimator, as well as its applicability in semiparametric causal inference with heterogeneous treatment effects. The finite-sample performance of our estimator is evaluated through Monte Carlo simulations. In an empirical setting related to microfinance program participation, we apply the new estimator to examine the average treatment effects and outcomes of counterfactual policies, and to propose an enhanced strategy for selecting the initial recipients of program information in social networks. |
| Date: | 2024–01 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2401.16275&r=big |
| By: | Phoebe Koundouri; Angelos Alamanos; Angelos Plataniotis; Charalampos Stavridis; Konstantinos Perifanos; Stathis Devves |
| Abstract: | The European Green Deal (EGD) is the growth strategy for Europe, covering multiple domains, and aiming to an equitable, climate neutral European Union by 2050. The UN Agenda 2030, encompassing 17 Sustainable Development Goals (SDGs), establishes the foundation for a global sustainability transition. The integration of the SDGs into the EGD is an overlooked issue in the literature, despite Europe's slow progress to achieve the sustainability targets. We employed a machine-learning text-mining method to evaluate the extent of SDG integration within the 74 EGD policy documents published during 2019�2023. The findings reveal a substantial alignment of EGD policies with SDGs related to clean energy (SDG7), climate action (SDG13), and sustainable consumption and production (SDG12). In contrast, there is a significant underrepresentation in areas related to social issues such as inequalities, poverty, hunger, health, education, gender equality, decent work, and peace, as indicated by lower alignment with SDGs 1, 2, 3, 4, 5, 8, 10, and 16. Temporal trends suggest a marginal increase in the attention given to environmental health (especially water and marine life) and gender equality. Furthermore, we illustrate the alignment of EGD policies with the six essential sustainability transformations proposed by the Sustainable Development Solutions Network (SDSN) in 2019 for the operationalization of the SDGs. The results indicate that besides the prevalence of "Energy Decarbonisation and Sustainable Industry", all areas have received attention, except for the "Health, Wellbeing and Demography". The findings call for a more integrated approach to address the complete spectrum of sustainability in a balanced manner. |
| Keywords: | European Green Dea, SDGs, Sustainability, Policy alignment, Text-mining, Machine Learning, Natural Language Processing, Sustainability Transformations |
| Date: | 2024–02–20 |
| URL: | http://d.repec.org/n?u=RePEc:aue:wpaper:2405&r=big |
| By: | Zhenglong Li; Vincent Tam; Kwan L. Yeung |
| Abstract: | Deep or reinforcement learning (RL) approaches have been adapted as reactive agents to quickly learn and respond with new investment strategies for portfolio management under the highly turbulent financial market environments in recent years. In many cases, due to the very complex correlations among various financial sectors, and the fluctuating trends in different financial markets, a deep or reinforcement learning based agent can be biased in maximising the total returns of the newly formulated investment portfolio while neglecting its potential risks under the turmoil of various market conditions in the global or regional sectors. Accordingly, a multi-agent and self-adaptive framework namely the MASA is proposed in which a sophisticated multi-agent reinforcement learning (RL) approach is adopted through two cooperating and reactive agents to carefully and dynamically balance the trade-off between the overall portfolio returns and their potential risks. Besides, a very flexible and proactive agent as the market observer is integrated into the MASA framework to provide some additional information on the estimated market trends as valuable feedbacks for multi-agent RL approach to quickly adapt to the ever-changing market conditions. The obtained empirical results clearly reveal the potential strengths of our proposed MASA framework based on the multi-agent RL approach against many well-known RL-based approaches on the challenging data sets of the CSI 300, Dow Jones Industrial Average and S&P 500 indexes over the past 10 years. More importantly, our proposed MASA framework shed lights on many possible directions for future investigation. |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2402.00515&r=big |
| By: | Bernhard Hientzsch |
| Abstract: | We consider two data driven approaches, Reinforcement Learning (RL) and Deep Trajectory-based Stochastic Optimal Control (DTSOC) for hedging a European call option without and with transaction cost according to a quadratic hedging P&L objective at maturity ("variance-optimal hedging" or "final quadratic hedging"). We study the performance of the two approaches under various market environments (modeled via the Black-Scholes and/or the log-normal SABR model) to understand their advantages and limitations. Without transaction costs and in the Black-Scholes model, both approaches match the performance of the variance-optimal Delta hedge. In the log-normal SABR model without transaction costs, they match the performance of the variance-optimal Barlett's Delta hedge. Agents trained on Black-Scholes trajectories with matching initial volatility but used on SABR trajectories match the performance of Bartlett's Delta hedge in average cost, but show substantially wider variance. To apply RL approaches to these problems, P&L at maturity is written as sum of step-wise contributions and variants of RL algorithms are implemented and used that minimize expectation of second moments of such sums. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2401.08600&r=big |
| By: | Tiziana De Stefano (Bank of Italy); Giuseppe Buscemi (Bank of Italy); Marco Fanari (Bank of Italy) |
| Abstract: | This paper explores the environmental, social and corporate governance (ESG) issues discussed during shareholder meetings held in 2021 and 2022 by major non-financial listed companies based in France, Germany and Italy. We examined shareholder meeting documentation in order to identify the questions, requests to supplement meeting agendas, draft resolutions and countermotions (forum rights) submitted by shareholders, as well as the responses provided by the companies. The analysis covers the entire sample of companies for 2021, while for 2022 it focuses on business sectors most exposed to climate transition risks and the effects of geopolitical tensions on energy supplies. For the 2021 shareholder meetings, the qualitative survey is complemented by a machine learning textual analysis (Latent Dirichlet Allocation, LDA) used as an aid to detect the most recurring themes addressed in the meetings. ESG factors were discussed in all shareholder meetings in the three markets examined for this study, albeit with varying emphasis. Shareholders mainly focused on the company’s environmental policy, including transition plans, alignment with international climate agreements, and the effects of business activities on the environment. As for social issues, the attention was largely on gender equality in the workforce, human rights, and the protection of health and wage conditions. As for corporate governance, there was a recurring interest in gender equality in management bodies, as well as in the opportunities to attend shareholder meetings remotely. |
| Keywords: | Shareholder rights; general meeting; corporate governance; Textual analysis; Latent Dirichlet Allocation |
| JEL: | C63 G30 G34 Q5 |
| Date: | 2024–02 |
| URL: | http://d.repec.org/n?u=RePEc:bdi:wpmisp:mip_045_24&r=big |
| By: | Daryna Grechyna |
| Abstract: | This paper evaluates the relative importance of natural and human factors in shaping public awareness of climate change. I compare the predictive efficacy of natural factors, represented by air temperature deviations from historical norms, and human factors, encompassing noteworthy political events focused on environmental policies and movements led by environmental activists, in forecasting the salience of climate change topic over weekly and annual horizons using regional European countries’ data. The salience of climate change is proxied by the Google search intensity data. The activists’ movements are measured by weekly Friday for Future strikes. The best-performing predictor in the short term (weeks), is the size of activists’ strikes and in the longer term (years), positive deviations of maximum air temperature from historical norms and political meetings focused on environmental policies. The inter-regional spatial relations, when taken into account, significantly improve the forecasts of the future public interest in climate change. |
| Keywords: | climate change, activists’ strikes, political meetings, weather |
| JEL: | Q01 Q52 Q58 C33 |
| Date: | 2024 |
| URL: | http://d.repec.org/n?u=RePEc:ces:ceswps:_10907&r=big |