|
on Computational Economics |
|
Issue of 2023‒12‒11
fourteen papers chosen by |
| By: | Koti S. Jaddu; Paul A. Bilokon |
| Abstract: | High-frequency trading is prevalent, where automated decisions must be made quickly to take advantage of price imbalances and patterns in price action that forecast near-future movements. While many algorithms have been explored and tested, analytical methods fail to harness the whole nature of the market environment by focusing on a limited domain. With the evergrowing machine learning field, many large-scale end-to-end studies on raw data have been successfully employed to increase the domain scope for profitable trading but are very difficult to replicate. Combining deep learning on the order books with reinforcement learning is one way of breaking down large-scale end-to-end learning into more manageable and lightweight components for reproducibility, suitable for retail trading. The following work focuses on forecasting returns across multiple horizons using order flow imbalance and training three temporal-difference learning models for five financial instruments to provide trading signals. The instruments used are two foreign exchange pairs (GBPUSD and EURUSD), two indices (DE40 and FTSE100), and one commodity (XAUUSD). The performances of these 15 agents are evaluated through backtesting simulation, and successful models proceed through to forward testing on a retail trading platform. The results prove potential but require further minimal modifications for consistently profitable trading to fully handle retail trading costs, slippage, and spread fluctuation. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.02088&r=cmp |
| By: | Reza Yarbakhsh; Mahdieh Soleymani Baghshah; Hamidreza Karimaghaie |
| Abstract: | Financial market forecasting remains a formidable challenge despite the surge in computational capabilities and machine learning advancements. While numerous studies have underscored the precision of computer-generated market predictions, many of these forecasts fail to yield profitable trading outcomes. This discrepancy often arises from the unpredictable nature of profit and loss ratios in the event of successful and unsuccessful predictions. In this study, we introduce a novel algorithm specifically designed for forecasting the profit and loss outcomes of trading activities. This is further augmented by an innovative approach for integrating these forecasts with previous predictions of market trends. This approach is designed for algorithmic trading, enabling traders to assess the profitability of each trade and calibrate the optimal trade size. Our findings indicate that this method significantly improves the performance of traditional trading strategies as well as algorithmic trading systems, offering a promising avenue for enhancing trading decisions. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.09148&r=cmp |
| By: | Abdel-Karim, Benjamin M.; Benlian, Alexander; Hinz, Oliver |
| Abstract: | Optimal investment decisions by institutional investors require accurate predictions with respect to the development of stock markets. Motivated by previous research that revealed the unsatisfactory performance of existing stock market prediction models, this study proposes a novel prediction approach. Our proposed system combines Artificial Intelligence (AI) with data from Virtual Investment Communities (VICs) and leverages VICs’ ability to support the process of predicting stock markets. An empirical study with two different models using real data shows the potential of the AI-based system with VICs information as an instrument for stock market predictions. VICs can be a valuable addition but our results indicate that this type of data is only helpful in certain market phases. |
| Date: | 2023–11–20 |
| URL: | http://d.repec.org/n?u=RePEc:dar:wpaper:141359&r=cmp |
| By: | Zengqing Wu; Run Peng; Xu Han; Shuyuan Zheng; Yixin Zhang; Chuan Xiao |
| Abstract: | Computer simulations offer a robust toolset for exploring complex systems across various disciplines. A particularly impactful approach within this realm is Agent-Based Modeling (ABM), which harnesses the interactions of individual agents to emulate intricate system dynamics. ABM's strength lies in its bottom-up methodology, illuminating emergent phenomena by modeling the behaviors of individual components of a system. Yet, ABM has its own set of challenges, notably its struggle with modeling natural language instructions and common sense in mathematical equations or rules. This paper seeks to transcend these boundaries by integrating Large Language Models (LLMs) like GPT into ABM. This amalgamation gives birth to a novel framework, Smart Agent-Based Modeling (SABM). Building upon the concept of smart agents -- entities characterized by their intelligence, adaptability, and computation ability -- we explore in the direction of utilizing LLM-powered agents to simulate real-world scenarios with increased nuance and realism. In this comprehensive exploration, we elucidate the state of the art of ABM, introduce SABM's potential and methodology, and present three case studies (source codes available at https://github.com/Roihn/SABM), demonstrating the SABM methodology and validating its effectiveness in modeling real-world systems. Furthermore, we cast a vision towards several aspects of the future of SABM, anticipating a broader horizon for its applications. Through this endeavor, we aspire to redefine the boundaries of computer simulations, enabling a more profound understanding of complex systems. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.06330&r=cmp |
| By: | Chunyang Huang; Shaoliang Zhang |
| Abstract: | This study presents an innovative method for predicting the market value of professional soccer players using explainable machine learning models. Using a dataset curated from the FIFA website, we employ an ensemble machine learning approach coupled with Shapley Additive exPlanations (SHAP) to provide detailed explanations of the models' predictions. The GBDT model achieves the highest mean R-Squared (0.8780) and the lowest mean Root Mean Squared Error (3, 221, 632.175), indicating its superior performance among the evaluated models. Our analysis reveals that specific skills such as ball control, short passing, finishing, interceptions, dribbling, and tackling are paramount within the skill dimension, whereas sprint speed and acceleration are critical in the fitness dimension, and reactions are preeminent in the cognitive dimension. Our results offer a more accurate, objective, and consistent framework for market value estimation, presenting useful insights for managerial decisions in player transfers. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.04599&r=cmp |
| By: | Yasuhiro Nakayama; Tomochika Sawaki |
| Abstract: | In this study, we have developed a dynamic asset allocation investment strategy using reinforcement learning techniques. To begin with, we have addressed the crucial issue of incorporating non-stationarity of financial time series data into reinforcement learning algorithms, which is a significant implementation in the application of reinforcement learning in investment strategies. Our findings highlight the significance of introducing certain variables such as regime change in the environment setting to enhance the prediction accuracy. Furthermore, the application of reinforcement learning in investment strategies provides a remarkable advantage of setting the optimization problem flexibly. This enables the integration of practical constraints faced by investors into the algorithm, resulting in efficient optimization. Our study has categorized the investment strategy formulation conditions into three main categories, including performance measurement indicators, portfolio management rules, and other constraints. We have evaluated the impact of incorporating these conditions into the environment and rewards in a reinforcement learning framework and examined how they influence investment behavior. |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:arx:papers:2311.04946&r=cmp |
| By: | Chen, Ying; Grith, Maria; Lai, Hannah L. H. |
| Abstract: | Implied volatility (IV) forecasting is inherently challenging due to its high dimensionality across various moneyness and maturity, and nonlinearity in both spatial and temporal aspects. We utilize implied volatility surfaces (IVS) to represent comprehensive spatial dependence and model the nonlinear temporal dependencies within a series of IVS. Leveraging advanced kernel-based machine learning techniques, we introduce the functional Neural Tangent Kernel (fNTK) estimator within the Nonlinear Functional Autoregression framework, specifically tailored to capture intricate relationships within implied volatilities. We establish the connection between fNTK and kernel regression, emphasizing its role in contemporary nonparametric statistical modeling. Empirically, we analyze S&P 500 Index options from January 2009 to December 2021, encompassing more than 6 million European calls and puts, thereby showcasing the superior forecast accuracy of fNTK.We demonstrate the significant economic value of having an accurate implied volatility forecaster within trading strategies. Notably, short delta-neutral straddle trading, supported by fNTK, achieves a Sharpe ratio ranging from 1.45 to 2.02, resulting in a relative enhancement in trading outcomes ranging from 77% to 583%. |
| Keywords: | Implied Volatility Surfaces; Neural Networks; Neural Tangent Kernel; Implied Volatility Forecasting; Nonlinear Functional Autoregression; Option Trading Strategies |
| JEL: | C14 C45 C58 G11 G13 G17 |
| Date: | 2023–10–24 |
| URL: | http://d.repec.org/n?u=RePEc:pra:mprapa:119022&r=cmp |
| By: | Tractenberg, Rochelle E. (Georgetown University) |
| Abstract: | Artificial Intelligence (AI) arises from computing and statistics, and as such, can be developed and deployed ethically when the ethical practice standards of each of these fields are followed. The Toronto Declaration was formulated in 2018 specifically to ensure that machine learning and AI could be held accountable for respecting, and promoting, universal human rights. The Code of Ethics and Professional Conduct of the Association of Computing Machinery (ACM, 2018) and the Ethical Guidelines for Statistical Practice of the American Statistical Association (ASA, 2022) describe the ethical practice standards for any person at any level of training or job title who utilizes computing (ACM) or statistical practices (ASA). These three reference documents can together define "what is ethical AI". All development, deployment, and use of computing is covered by the ACM Code; the ASA defines statistical practice to "include activities such as: designing the collection of, summarizing, processing, analyzing, interpreting, or presenting, data; as well as model or algorithm development and deployment.” Just as the Toronto Declaration describes universal human rights protections, the ACM and ASA ethical practice standards apply to professionals, individuals with diverse background or jobs that include computing and statistical practices at any point, and employers, clients, organizations, and institutions that employ or utilize the outputs from computing and statistical practices worldwide. The ACM Code of Ethics has four Principles, including one specifically for Leaders with seven elements. The ASA Ethical Guidelines include eight principles and an Appendix; one Guideline Principle (G. Responsibilities of Leaders, Supervisors, and Mentors in Statistical Practice) with its five elements and the Appendix (Responsibilities of organizations/institutions) with its 12 elements are specifically intended to support workplace engagement with, and support of, ethical statistical practices, plus, the specific roles and responsibilities of those in leadership positions. These ethical practice standards can support both individual practitioners', and leaders', meeting their obligations for ethical AI worldwide. |
| Date: | 2023–11–13 |
| URL: | http://d.repec.org/n?u=RePEc:osf:socarx:8e6pv&r=cmp |
| By: | Mathias Kraus (University of Erlangen-Nuremberg); Julia Bingler (University of Oxford); Markus Leippold (University of Zurich; Swiss Finance Institute); Tobias Schimanski (University of Zurich); Chiara Colesanti Senni (ETH Zürich; University of Zurich); Dominik Stammbach (ETH Zürich); Saeid Vaghefi (University of Zurich); Nicolas Webersinke (Friedrich-Alexander-Universität Erlangen-Nürnberg) |
| Abstract: | Large language models (LLMs) have significantly transformed the landscape of artificial intelligence by demonstrating their ability to generate human-like text across diverse topics. However, despite their impressive capabilities, LLMs lack recent information and often employ imprecise language, which can be detrimental in domains where accuracy is crucial, such as climate change. In this study, we make use of recent ideas to harness the potential of LLMs by viewing them as agents that access multiple sources, including databases containing recent and precise information about organizations, institutions, and companies. We demonstrate the effectiveness of our method through a prototype agent that retrieves emission data from ClimateWatch (https://www.climatewatchdata.org/) and leverages general Google search. By integrating these resources with LLMs, our approach overcomes the limitations associated with imprecise language and delivers more reliable and accurate information in the critical domain of climate change. This work paves the way for future advancements in LLMs and their application in domains where precision is of paramount importance. |
| Date: | 2023–10 |
| URL: | http://d.repec.org/n?u=RePEc:chf:rpseri:rp2399&r=cmp |
| By: | Nils Grashof; Alexander Kopka |
| Abstract: | Artificial intelligence (AI) is seen as a key technology for economic growth. However, the impact of AI on firm productivity has been under researched – particularly through the lens of inequality and clusters. Based on a unique sample of German firms, filling at least one patent between 2013 and 2019, we find evidence for a positive influence of AI on firm productivity. Moreover, our analysis shows that while AI knowledge does not contribute to productivity divergences in general, it increases the productivity gap between laggard and all other firms. Nevertheless, this effect is reduced through the localisation in clusters. |
| Keywords: | Artificial intelligence, Inequality, Productivity, Clusters, Patents, Firm-level |
| JEL: | O18 O30 R10 |
| Date: | 2023–11 |
| URL: | http://d.repec.org/n?u=RePEc:atv:wpaper:2304&r=cmp |
| By: | Morteza Alaeddini (AUT - Amirkabir University of Technology, UGA - Université Grenoble Alpes, CERAG - Centre d'études et de recherches appliquées à la gestion - UGA - Université Grenoble Alpes); Julie Dugdale (LIG - Laboratoire d'Informatique de Grenoble - CNRS - Centre National de la Recherche Scientifique - UGA - Université Grenoble Alpes - Grenoble INP - Institut polytechnique de Grenoble - Grenoble Institute of Technology - UGA - Université Grenoble Alpes); Paul Reaidy (CERAG - Centre d'études et de recherches appliquées à la gestion - UGA - Université Grenoble Alpes); Philippe Madiès (CERAG - Centre d'études et de recherches appliquées à la gestion - UGA - Université Grenoble Alpes) |
| Abstract: | Trust is crucial in economic complex adaptive systems, where agents frequently change the other side of their interactions, which often leads to changes in the system's structure. In such a system, agents who seek as much as possible to build lasting trust relationships for long-term confident interactions with their counterparts decide whom to interact with based on their level of trust in existing partners. A trust crisis refers to the time when the level of trust between agents drops so much that there is no incentive to interact, a situation that ultimately leads to the collapse of the system. This paper presents an agent-based model of the interbank market and evaluates the effects of using a voting-based consensus mechanism embedded in a blockchain-based loan system on maintaining trust between agents and system stability. In this paper, we rely on the fact that blockchain as a distributed system only manages the existing trust and does not create it on its own. Furthermore, this study uses actual blockchain technology in its simulation rather than simply presenting an abstraction. |
| Keywords: | Agent-based simulation, Asymmetric information, Confidence, Distributed ledger, Interbank call loan market, Uncertainty |
| Date: | 2023–09–30 |
| URL: | http://d.repec.org/n?u=RePEc:hal:journl:hal-04266077&r=cmp |
| By: | George Athanasopoulos; Rob J Hyndman; Raffaele Mattera |
| Abstract: | This paper discusses the use of forecast reconciliation with stock price time series and the corresponding stock index. The individual stock price series may be grouped using known meta-data or other clustering methods. We propose a novel forecasting framework that combines forecast reconciliation and clustering, to lead to better forecasts of both the index and the individual stock price series. The proposed approach is applied to the Dow Jones Industrial Average Index and its component stocks. The results demonstrate empirically that reconciliation improves forecasts of the stock market index and its constituents. |
| Keywords: | financial time series, hierarchical forecasting, clustering, unsupervised learning, prediction, machine learning, finance |
| JEL: | C53 C10 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:msh:ebswps:2023-17&r=cmp |
| By: | Ikumo Isono (Economic Research Institute for ASEAN and East Asia (ERIA)); Hilmy Prilliadi (Economic Research Institute for ASEAN and East Asia (ERIA)) |
| Abstract: | Artificial intelligence (AI) is attracting significant attention worldwide in 2023 because of its potential to transform economies and societies. The Association of Southeast Asian Nations (ASEAN) must accelerate the debate on AI for five compelling reasons. First, narrowing the gaps in AI readiness within ASEAN is essential to share the benefits of AI equitably. Second, there are concerns that rapid advances in AI could result in job loss, and retraining is needed. Third, AI systems must be developed from an ASEAN-centric perspective to overcome prejudice and align AI with ASEAN values. Fourth, as developed countries implement AI regulations, ASEAN needs to consider the need for its own regional policies. Finally, now is the perfect time to discuss the positioning of AI in the regional framework as ASEAN’s digital integration initiative progresses. The paper discusses the significance of AI in 2023, the challenges in ASEAN, the need for its own policies, and policy recommendations. |
| Keywords: | Artificial Intelligence; ASEAN; Employment; Regulation; Ethics |
| JEL: | D78 F15 K23 O33 O38 |
| Date: | 2023–11–03 |
| URL: | http://d.repec.org/n?u=RePEc:era:wpaper:dp-2023-16&r=cmp |
| By: | Angelika Abramiuk-Szurlej; Mikolaj Szurlej; Katarzyna Sznajd-Weron |
| Abstract: | Agent-based modeling (ABM), a bottom-up stochastic approach for simulating the interactions of multiple autonomous agents, is gaining popularity in the field of managing pro-environmental behavior change. In the field of ecology, it is a well-established and rigorous scientific method. However, within the social sciences, it is often criticized for its lack of rigor. In this paper, we demonstrate how best practices from ABM in ecology can be applied to the study of pro-environmental social change, with a special focus on energy-related behaviors. We argue that the two stages of ABM, namely description and verification, are fundamental for establishing ABM as a rigorous research method. However, upon critically reviewing the existing literature on ABM of energy-related behaviors, we find that these stages are frequently absent or poorly executed. Therefore, we provide a practical illustration of how to effectively execute these stages using an example of a model introduced in 2016 to study the diffusion of green products and practices. We describe the model using the ODD (Overview, Design concepts, Details) protocol. Furthermore, we present two different approaches to model analysis borrowed from the theory of complex systems to ensure rigorous model verification. We also clarify the circumstances under which the agent-based model can be reduced to an analytical model and when such reduction is not feasible. |
| Keywords: | agent-based model; ODD protocol; pro-environmental behavior change |
| JEL: | C63 D81 D91 O33 O35 Q42 Q5 |
| Date: | 2023 |
| URL: | http://d.repec.org/n?u=RePEc:ahh:wpaper:worms2302&r=cmp |