|
on Computational Economics |
| By: | Venkat Ram Reddy Ganuthula; Krishna Kumar Balaraman |
| Abstract: | As artificial intelligence becomes increasingly integrated into professional and personal domains, traditional metrics of human intelligence require reconceptualization. This paper introduces the Artificial Intelligence Quotient (AIQ), a novel measurement framework designed to assess an individual's capacity to effectively collaborate with and leverage AI systems, particularly Large Language Models (LLMs). Building upon established cognitive assessment methodologies and contemporary AI interaction research, we present a comprehensive framework for quantifying human-AI collaborative intelligence. This work addresses the growing need for standardized evaluation of AI-augmented cognitive capabilities in educational and professional contexts. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2503.16438 |
| By: | Magaletti, Nicola; Notarnicola, Valeria; Di Molfetta, Mauro; Mariani, Stefano; Leogrande, Angelo |
| Abstract: | The paper investigates the deployment of data analytics and machine learning to improve welding quality in Tecnomulipast srl, a small-to-medium sized manufacturing firm located in Puglia, Italy. The firm produces food machine components and more recently mechanized its laser welding process with the introduction of an IoT-enabled system integrating photographic control. The investment, underwritten by the Apulia Region under PIA (Programmi Integrati di Agevolazione) allowed Tecnomulipast to not only mechanize its production line but also embark upon wider digital transformation. This involved the creation of internal data analytics infrastructures that have the capability to underpin machine learning and artificial intelligence applications. This paper addresses a prediction of weld bead width (LC) with a dataset of 1, 000 observations. Input variables are laser power (PL), pulse time (DI), frequency (FI), beam diameter (DF), focal position (PF), travel speed (VE), trajectory accuracy (TR), laser angle (AN), gas flow (FG), gas purity (PG), ambient temperature (TE), and penetration depth (PE). The parameters were exploited to build and validate some supervised machine learning algorithms like Decision Trees, Random Forest, K-Nearest Neighbors, Support Vector Machines, Neural Networks, and Linear Regression. The performance of the models was measured by MSE, RMSE, MAE, MAPE, and R². Ensemble methods like Random Forest and Boosting performed the highest. Feature importance analysis determined that laser power, gas flow, and trajectory accuracy are the key variables. This project showcases the manner in which Tecnomulipast has benefited from public investment to introduce digital transformation and adopt data-driven strategies within Industry 4.0. |
| Keywords: | Tecnomulipast, laser welding, machine learning, digital transformation, Industry 4.0. |
| JEL: | C45 C5 C53 L23 O33 |
| Date: | 2025–04–24 |
| URL: | https://d.repec.org/n?u=RePEc:pra:mprapa:124548 |
| By: | Giovanni Ballarin; Jacopo Capra; Petros Dellaportas |
| Abstract: | Stock return prediction is a problem that has received much attention in the finance literature. In recent years, sophisticated machine learning methods have been shown to perform significantly better than ''classical'' prediction techniques. One downside of these approaches is that they are often very expensive to implement, for both training and inference, because of their high complexity. We propose a return prediction framework for intraday returns at multiple horizons based on Echo State Network (ESN) models, wherein a large portion of parameters are drawn at random and never trained. We show that this approach enjoys the benefits of recurrent neural network expressivity, inherently efficient implementation, and strong forecasting performance. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.19623 |
| By: | Ziqi Li |
| Abstract: | This chapter discusses the opportunities of eXplainable Artificial Intelligence (XAI) within the realm of spatial analysis. A key objective in spatial analysis is to model spatial relationships and infer spatial processes to generate knowledge from spatial data, which has been largely based on spatial statistical methods. More recently, machine learning offers scalable and flexible approaches that complement traditional methods and has been increasingly applied in spatial data science. Despite its advantages, machine learning is often criticized for being a black box, which limits our understanding of model behavior and output. Recognizing this limitation, XAI has emerged as a pivotal field in AI that provides methods to explain the output of machine learning models to enhance transparency and understanding. These methods are crucial for model diagnosis, bias detection, and ensuring the reliability of results obtained from machine learning models. This chapter introduces key concepts and methods in XAI with a focus on Shapley value-based approaches, which is arguably the most popular XAI method, and their integration with spatial analysis. An empirical example of county-level voting behaviors in the 2020 Presidential election is presented to demonstrate the use of Shapley values and spatial analysis with a comparison to multi-scale geographically weighted regression. The chapter concludes with a discussion on the challenges and limitations of current XAI techniques and proposes new directions. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.00591 |
| By: | Lars Hornuf; David J. Streich; Niklas Töllich |
| Abstract: | Retrieval-augmented generation (RAG) has emerged as a promising way to improve task-specific performance in generative artificial intelligence (GenAI) applications such as large language models (LLMs). In this study, we evaluate the performance implications of providing various types of domain-specific information to LLMs in a simple portfolio allocation task. We compare the recommendations of seven state-of-the-art LLMs in various experimental conditions against a benchmark of professional financial advisors. Our main result is that the provision of domain-specific information does not unambiguously improve the quality of recommendations. In particular, we find that LLM recommendations underperform recommendations by human financial advisors in the baseline condition. However, providing firm-specific information improves historical performance in LLM portfolios and closes the gap with human advisors. Performance improvements are achieved through higher exposure to market risk and not through an increase in mean-variance efficiency within the risky portfolio share. Notably, portfolio risk increases primarily for risk-averse investors. We also document that quantitative firm-specific information affects recommendations more than qualitative firm-specific information, and that equipping models with generic finance theory does not affect recommendations. |
| Keywords: | generative artificial intelligence, large language models, domain-specific information, retrieval-augmented generation, portfolio management, portfolio allocation. |
| JEL: | G00 G11 G40 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:ces:ceswps:_11862 |
| By: | Cheng Wang; Chuwen Wang; Shirong Zeng; Changjun Jiang |
| Abstract: | After decades of evolution, the financial system has increasingly deviated from an idealized framework based on theorems. It necessitates accurate projections of complex market dynamics and human behavioral patterns. With the development of data science and machine intelligence, researchers are trying to digitalize and automate market prediction. However, existing methodologies struggle to represent the diversity of individuals and are regardless of the domino effects of interactions on market dynamics, leading to the poor performance facing abnormal market conditions where non-quantitative information dominates the market. To alleviate these disadvantages requires the introduction of knowledge about how non-quantitative information, like news and policy, affects market dynamics. This study investigates overcoming these challenges through rehearsing potential market trends based on the financial large language model agents whose behaviors are aligned with their cognition and analyses in markets. We propose a hierarchical knowledge architecture for financial large language model agents, integrating fine-tuned language models and specialized generators optimized for trading scenarios. For financial market, we develop an advanced interactive behavioral simulation system that enables users to configure agents and automate market simulations. In this work, we take commodity futures as an example to research the effectiveness of our methodologies. Our real-world case simulation succeeds in rehearsing abnormal market dynamics under geopolitical events and reaches an average accuracy of 3.4% across various points in time after the event on predicting futures price. Experimental results demonstrate our method effectively leverages diverse information to simulate behaviors and their impact on market dynamics through systematic interaction. |
| Date: | 2025–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2503.20787 |
| By: | Castro-Iragorri, Carlos (Universidad del Rosario); Parra-Diaz, Manuel (Universidad del Rosario) |
| Abstract: | Recent advances in deep learning have spurred the development of end-to-end frameworks for portfolio optimization that utilize implicit layers. However, many such implementations are highly sensitive to neural network initialization, undermining performance consistency. This research introduces a robust end-to-end framework tailored for risk budgeting portfolios that effectively reduces sensitivity to initialization. Importantly, this enhanced stability does not compromise portfolio performance, as our framework consistently outperforms the risk parity benchmark. |
| Keywords: | end-to-end framework; neural networks; risk budgeting; stability |
| JEL: | C13 C45 G11 |
| Date: | 2025–03–05 |
| URL: | https://d.repec.org/n?u=RePEc:col:000092:021367 |
| By: | Manuel Parra-Diaz; Carlos Castro-Iragorri |
| Abstract: | Recent advances in deep learning have spurred the development of end-to-end frameworks for portfolio optimization that utilize implicit layers. However, many such implementations are highly sensitive to neural network initialization, undermining performance consistency. This research introduces a robust end-to-end framework tailored for risk budgeting portfolios that effectively reduces sensitivity to initialization. Importantly, this enhanced stability does not compromise portfolio performance, as our framework consistently outperforms the risk parity benchmark. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.19980 |
| By: | Tiantian Tu |
| Abstract: | Time series forecasting is crucial for decision-making across various domains, particularly in financial markets where stock prices exhibit complex and non-linear behaviors. Accurately predicting future price movements is challenging due to the difficulty of capturing both short-term fluctuations and long-term dependencies in the data. Convolutional Neural Networks (CNNs) are well-suited for modeling localized, short-term patterns but struggle with long-range dependencies due to their limited receptive field. In contrast, Transformers are highly effective at capturing global temporal relationships and modeling long-term trends. In this paper, we propose a hybrid architecture that combines CNNs and Transformers to effectively model both short- and long-term dependencies in financial time series data. We apply this approach to forecast stock price movements for S\&P 500 constituents and demonstrate that our model outperforms traditional statistical models and popular deep learning methods in intraday stock price forecasting, providing a robust framework for financial prediction. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.19309 |
| By: | Marco Zanotti |
| Abstract: | In an era of increasing computational capabilities and growing environmental consciousness, organizations face a critical challenge in balancing the accuracy of their forecasting models with computational efficiency and sustainability. Global forecasting models, which leverage data across multiple time series to improve prediction accuracy, lowering the computational time, have gained significant attention over the years. However, the common practice of retraining these models with new observations raises important questions about the costs of producing forecasts. Using ten different machine learning and deep learning models, we analyzed various retraining scenarios, ranging from continuous updates to no retraining at all, across two large retail datasets. We showed that less frequent retraining strategies can maintain the forecast accuracy while reducing the computational costs, providing a more sustainable approach to large-scale forecasting. We also found that machine learning models are a marginally better choice to reduce the costs of forecasting when coupled with less frequent model retraining strategies as the frequency of the data increases. Our findings challenge the conventional belief that frequent retraining is essential for maintaining forecasting accuracy. Instead, periodic retraining offers a good balance between predictive performance and efficiency, both in the case of point and probabilistic forecasting. These insights provide actionable guidelines for organizations seeking to optimize forecasting pipelines while reducing costs and energy consumption. |
| Keywords: | Time series, Demand forecasting, Forecasting competitions, Cross-learning, Global models, Machine learning, Deep learning, Green AI, Conformal predictions. |
| JEL: | C53 C52 C55 |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:mib:wpaper:551 |
| By: | Lawrence, Alice |
| Abstract: | In an era marked by increasing global uncertainties from pandemics and geopolitical tensions to climate-related disruptions, supply chain resilience has emerged as a strategic imperative. This paper explores the transformative role of Artificial Intelligence (AI) in enhancing the adaptability, visibility, and responsiveness of modern supply chains. By delving into real-world case studies and recent advancements, we analyze how AI-powered tools such as predictive analytics, intelligent automation, and machine learning enable companies to anticipate disruptions, optimize logistics, and accelerate decision-making. Rather than merely reacting to crises, AI equips organizations with the foresight and agility needed to proactively manage risks and sustain operations. While the integration of AI is not without challenges including data privacy concerns, integration complexities, and workforce readiness, this study underscores its potential as a cornerstone technology for future-proofing global supply networks. Ultimately, we argue that AI is not just a tool for efficiency, but a critical enabler of resilience in a volatile world. Keywords: Supply Chain Resilience, Artificial Intelligence, Predictive Analytics, Risk Management, Intelligent Automation, Machine Learning, Logistics Optimization, Disruption Forecasting, Digital Transformation, Strategic Agility |
| Date: | 2024–07–11 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:4x2jz_v1 |
| By: | Amelie BARBIER-GAUCHARD; Emmanouil SOFIANOS |
| Abstract: | The situation of public finance in the eurozone remains a burning issue for certain Euro area countries. The financial markets, the main lenders of the Member States, are more attentive than ever to any factor which could affect the trajectory of public debt in the long term. The risk of bankruptcy of a Member State and a domino effect for the entire monetary union represents the ultimate risk weighing on the Eurozone. This paper aims to forecast the public debt, with a universal model, on a national level within the Euro area. We use a dataset that includes 566 independent variables (economic, financial, institutional, political and social) for 17 Euro area countries, spanning the period from 2000 to 2022 in annual frequency. The dataset is fed to four machine learning (ML) algorithms: Decision Trees, Random Forests, XGBoost and Support Vector Machines (SVM). We also employ the Elastic-Net Regression algorithm from the area of Econometrics. The best model is an XGBoost with an out-of-sample MAPE of 8.41%. Moreover, it outperforms the projections of European Commission and IMF. According to the VIM from XGBoost, the most influential variables are the past values of public debt, the male population in the ages 50-54, the regulatory quality, the control of corruption, the female employment to population ratio for the ages over 15 and the 10 year bond spread. |
| Keywords: | Public Debt; Euro Area; Machine Learning; Forecasting. |
| JEL: | C53 H63 H68 |
| Date: | 2024 |
| URL: | https://d.repec.org/n?u=RePEc:ulp:sbbeta:2024-47 |
| By: | Alina Landowska; Robert A. K{\l}opotek; Dariusz Filip; Konrad Raczkowski |
| Abstract: | This study examines the relationship between GDP growth and Gross Fixed Capital Formation (GFCF) across developed economies (G7, EU-15, OECD) and emerging markets (BRICS). We integrate Random Forest machine learning (non-linear regression) with traditional econometric models (linear regression) to better capture non-linear interactions in investment analysis. Our findings reveal that while GDP growth positively influences corporate investment, its impact varies significantly by region. Developed economies show stronger GDP-GFCF linkages due to stable financial systems, while emerging markets demonstrate weaker connections due to economic heterogeneity and structural constraints. Random Forest models indicate that GDP growth's importance is lower than suggested by traditional econometrics, with lagged GFCF emerging as the dominant predictor-confirming investment follows path-dependent patterns rather than short-term GDP fluctuations. Regional variations in investment drivers are substantial: taxation significantly influences developed economies but minimally affects BRICS, while unemployment strongly drives investment in BRICS but less so elsewhere. We introduce a parallelized p-value importance algorithm for Random Forest that enhances computational efficiency while maintaining statistical rigor through sequential testing methods (SPRT and SAPT). The research demonstrates that hybrid methodologies combining machine learning with econometric techniques provide more nuanced understanding of investment dynamics, supporting region-specific policy design and improving forecasting accuracy. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.20993 |
| By: | Saizhuo Wang; Hao Kong; Jiadong Guo; Fengrui Hua; Yiyan Qi; Wanyun Zhou; Jiahao Zheng; Xinyu Wang; Lionel M. Ni; Jian Guo |
| Abstract: | The field of artificial intelligence (AI) in quantitative investment has seen significant advancements, yet it lacks a standardized benchmark aligned with industry practices. This gap hinders research progress and limits the practical application of academic innovations. We present QuantBench, an industrial-grade benchmark platform designed to address this critical need. QuantBench offers three key strengths: (1) standardization that aligns with quantitative investment industry practices, (2) flexibility to integrate various AI algorithms, and (3) full-pipeline coverage of the entire quantitative investment process. Our empirical studies using QuantBench reveal some critical research directions, including the need for continual learning to address distribution shifts, improved methods for modeling relational financial data, and more robust approaches to mitigate overfitting in low signal-to-noise environments. By providing a common ground for evaluation and fostering collaboration between researchers and practitioners, QuantBench aims to accelerate progress in AI for quantitative investment, similar to the impact of benchmark platforms in computer vision and natural language processing. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.18600 |
| By: | Anna P. Kwossek; David J. Pr\"omel; Josef Teichmann |
| Abstract: | We identify various classes of neural networks that are able to approximate continuous functions locally uniformly subject to fixed global linear growth constraints. For such neural networks the associated neural stochastic differential equations can approximate general stochastic differential equations, both of It\^o diffusion type, arbitrarily well. Moreover, quantitative error estimates are derived for stochastic differential equations with sufficiently regular coefficients. |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2503.16696 |
| By: | Hirmer, Christine (Institute for Employment Research (IAB), Nuremberg, Germany); Metzger, Lina-Jeanette (Institute for Employment Research (IAB), Nuremberg, Germany) |
| Abstract: | "The constantly growing amount of digitally accessible text data and advances in natural language processing (NLP) have made text mining a key technology. The "Temi-Box" is a modular construction kit designed for text mining applications. It enables automated text classification, topic categorization, and clustering without necessitating extensive programming expertise. Developed on the basis of the keywording and topic assignment of publications for the IAB Info Platform and financed by EU funds, it is available as an open source project. This research report documents the development and application of the Temi-Box and illustrates its use and the interpretation of the results obtained. Text mining extracts knowledge from unstructured texts using methods such as classification and clustering. The modular Temi-Box provides users with established methods in a user-friendly way and supports users with a pipeline architecture that simplifies standardised processes such as data preparation and model training. It incorporates both current and traditional approaches to text representation, such as BERT and TF-IDF, and offers a variety of algorithms for text classification and clustering, including K-Nearest Neighbors (KNN), binary and multinomial classifiers as layers in neural networks and K-Means. Various evaluation metrics facilitate the assessment of model performance and the comparison of different approaches. Experiments on automated topic assignment and the identification of key topics illustrate the use of the Temi-Box and the interpretation of the results. Based on a dataset with 1, 932 IAB publications and 105 topics, the results show that BERT-based models, such as GermanBERT, consistently achieve the best results. Binary classifiers prove to be particularly flexible and accurate, while TF-IDF-based approaches offer robust alternatives with less complexity. Clustering remains a challenge, especially when content overlaps. The Temi-Box is a highly versatile instrument. In addition to the application for the IAB Info Platform described in this research report, it can be used in numerous areas, such as the analysis of job advertisements, job and company descriptions, keywording of publications or for sentiment analysis. It can also be extended for use in question-and-answer systems or for named entity recognition. The Temi-Box facilitates the application of text mining methods for a broad user base and offers numerous customization options. It reduces the effort involved in developing and comparing models. Its open source availability promotes the further development and integration of the Temi-Box into various research projects. This enables users to adapt the platform to specific needs and integrate new functions. The report shows the potential of the Temi-Box to advance the digitization and automation of text data analysis. At the same time, challenges such as ensuring data quality and the interpretability of the models remain. These aspects require continuous validation and further development in order to further improve the effectiveness and reliability of text mining methods." (Author's abstract, IAB-Doku) ((en)) |
| Keywords: | Bundesrepublik Deutschland ; IAB-Open-Access-Publikation ; Natural Language Processing ; Automatisierung ; Datenanalyse ; IAB ; Indexierung ; Klassifikation ; Anwendung ; Text Mining ; Veröffentlichung |
| Date: | 2025–05–08 |
| URL: | https://d.repec.org/n?u=RePEc:iab:iabfob:202513 |
| By: | Joao Felipe Gueiros; Hemanth Chandravamsi; Steven H. Frankel |
| Abstract: | This paper explores the use of deep residual networks for pricing European options on Petrobras, one of the world's largest oil and gas producers, and compares its performance with the Black-Scholes (BS) model. Using eight years of historical data from B3 (Brazilian Stock Exchange) collected via web scraping, a deep learning model was trained using a custom built hybrid loss function that incorporates market data and analytical pricing. The data for training and testing were drawn between the period spanning November 2016 to January 2025, using an 80-20 train-test split. The test set consisted of data from the final three months: November, December, and January 2025. The deep residual network model achieved a 64.3\% reduction in the mean absolute error for the 3-19 BRL (Brazilian Real) range when compared to the Black-Scholes model on the test set. Furthermore, unlike the Black-Scholes solution, which tends to decrease its accuracy for longer periods of time, the deep learning model performed accurately for longer expiration periods. These findings highlight the potential of deep learning in financial modeling, with future work focusing on specialized models for different price ranges. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.20088 |
| By: | Jennifer Priefer Author-1-Name-First: Jennifer Author-1-Name-Last: Priefer (Paderborn University); Jan-Peter Kucklick Author-2-Name-First: Jan-Peter Author-2-Name-Last: Kucklick (Paderborn University); Daniel Beverungen Author-3-Name-First: Daniel Author-3-Name-Last: Beverungen (Paderborn University); Oliver Müller Author-3-Name-First: Oliver Author-3-Name-Last: Müller (Paderborn University) |
| Abstract: | Information systems have proven their value in facilitating pricing decisions. Still, predicting prices for complex goods, such as houses, remains challenging due to information asymmetries that obscure their qualities. Beyond search qualities that sellers can identify before a purchase, complex goods also possess experience qualities only identifiable ex-post. While research has discussed how information asymmetries cause market failure, it remains unclear how information systems can account for search and experience qualities of complex goods to enable their pricing in online markets. In a machine learning-based study, we quantify their predictive power for online real estate pricing, using geographic information systems and computer vision to incorporate spatial and image data into price prediction. We find that leveraging these secondary use data can transform some experience qualities into search qualities, increasing predictive power by up to 15.4%. We conclude that spatial and image data can provide valuable resources for improving price predictions for complex goods. |
| Keywords: | information asymmetries; real estate appraisal; SEC theory; machine learning; geographic information systems; computer vision |
| JEL: | C53 D82 R31 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:pdn:dispap:138 |
| By: | Ngueuleweu Tiwang Gildas |
| Abstract: | Machine learning detects patterns, block chain guarantees trust and immutability, and modern causal inference identifies directional linkages, yet none alone exposes the full energetic anatomy of complex systems; the Hamiltonian Higher Order Elasticity Dynamics(2HOED) framework bridges these gaps. Grounded in classical mechanics but extended to Economics order elasticity terms, 2HOED represents economic, social, and physical systems as energy-based Hamiltonians whose position, velocity, acceleration, and jerk of elasticity jointly determine systemic power, Inertia, policy sensitivity, and marginal responses. Because the formalism is scaling free and coordinate agnostic, it transfers seamlessly from financial markets to climate science, from supply chain logistics to epidemiology, thus any discipline in which adaptation and shocks coexist. By embedding standard econometric variables inside a Hamiltonian, 2HOED enriches conventional economic analysis with rigorous diagnostics of resilience, tipping points, and feedback loops, revealing failure modes invisible to linear models. Wavelet spectra, phase space attractors, and topological persistence diagrams derived from 2HOED expose multistage policy leverage that machine learning detects only empirically and block chain secures only after the fact. For economists, physicians and other scientists, the method opens a new causal energetic channel linking biological or mechanical elasticity to macro level outcomes. Portable, interpretable, and computationally light, 2HOED turns data streams into dynamical energy maps, empowering decision makers to anticipate crises, design adaptive policies, and engineer robust systems delivering the predictive punch of AI with the explanatory clarity of physics. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.21062 |
| By: | Srivastava, Varad |
| Abstract: | Traditional banking, which relies on relationship managers to provide personalized financial advice, recommend products and assess risk, has been left relatively untouched from potential enhancements of Generative AI. In this work, we propose - EAGLE, a multi-agent system for this task, which automates banking recommendations while incorporating real-time, risk-aware financial planning and augments and enhances operations by cutting down on time taken for research on customer profiles, products, financial plans as well as call handling. Through simulated experiments and novel proposed metrics, we establish robust performance on this task using our framework. Our proposed multi-agent system enhances personalization of products recommendation, enables risk-aware financial planning and asset allocation, as well as establishes a foundation for next-generation banking systems. |
| Date: | 2025–04–28 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:dwspv_v1 |
| By: | Grané Chávez, Aurea; Scielzo Ortiz, Fabio |
| Abstract: | In this work new robust efficient clustering algorithms for large datasets of mixedtype data are proposed and implemented in a new Python package called FastKmedoids. Their performance is analyzed through an extensive simulation study, and compared to a wide range of existing clustering alternatives in terms of both predictive power and computational efficiency. MDS is used to visualize clustering results. |
| Keywords: | Clustering; Fast k-medoids; Outliers; Robust mahalanobis; Clustering; Fast K-Medoids; Generalized Gower; Multivariate Heterogeneous Data; Outliers; Robust Mahalanobis; Generalized Gower; Multivariate heterogeneous data |
| Date: | 2025–05–12 |
| URL: | https://d.repec.org/n?u=RePEc:cte:wsrepe:46673 |
| By: | Chris Santos-Lang; Christopher M. Homan |
| Abstract: | This paper contributes a new way to evaluate AI. Much as one might evaluate a machine in terms of its performance at chess, this approach involves evaluating a machine in terms of its performance at a game called "MAD Chairs". At the time of writing, evaluation with this game exposed opportunities to improve Claude, Gemini, ChatGPT, Qwen and DeepSeek. Furthermore, this paper sets a stage for future innovation in game theory and AI safety by providing an example of success with non-standard approaches to each: studying a game beyond the scope of previous game theoretic tools and mitigating a serious AI safety risk in a way that requires neither determination of values nor their enforcement. |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2503.20986 |
| By: | Michael S. Barr |
| Date: | 2025–05–09 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedgsq:99947 |
| By: | Ricardo Dahis (Department of Economics, Monash University); Martin Mattsson (Department of Economics, National University of Singapore); Nathalia Sales (Department of Economics, PUC-Rio) |
| Abstract: | We revisit the literature about the impact of reelection incentives on corruption with an extended dataset of corruption audit reports classified with Large Language Model (LLM). We first show that correlations between the LLM-generated corruption measures and manually coded assessments are comparable to correlations among the manual datasets themselves. Our results support previous findings in the literature, although the result is only statistically significant for one out of three measures of corruption. We document significant heterogeneity in the effect over time and investigate several explanations for these empirical patterns, including changing composition of politicians and increasing probability of legal penalties. |
| Keywords: | reelection incentives, corruption, LLM |
| JEL: | D72 K42 O17 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:mos:moswps:2025-08 |
| By: | Robert Millar; Jinglai Li |
| Abstract: | Optimal portfolio allocation is often formulated as a constrained risk problem, where one aims to minimize a risk measure subject to some performance constraints. This paper presents new Bayesian Optimization algorithms for such constrained minimization problems, seeking to minimize the conditional value-at-risk (a computationally intensive risk measure) under a minimum expected return constraint. The proposed algorithms utilize a new acquisition function, which drives sampling towards the optimal region. Additionally, a new two-stage procedure is developed, which significantly reduces the number of evaluations of the expensive-to-evaluate objective function. The proposed algorithm's competitive performance is demonstrated through practical examples. |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2503.17737 |
| By: | Eduardo Abi Jaber; Elie Attal |
| Abstract: | We introduce a novel simulation scheme, iVi (integrated Volterra implicit), for integrated Volterra square-root processes and Volterra Heston models based on the Inverse Gaussian distribution. The scheme is designed to handle $L^1$ kernels with singularities by relying solely on integrated kernel quantities, and it preserves the non-decreasing property of the integrated process. We establish weak convergence of the iVi scheme by reformulating it as a stochastic Volterra equation with a measure kernel and proving a stability result for this class of equations. Numerical results demonstrate that convergence is achieved with very few time steps. Remarkably, for the rough fractional kernel, unlike existing schemes, convergence seems to improve as the Hurst index $H$ decreases and approaches $-1/2$. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.19885 |
| By: | Mr. Tanai Khiaonarong; Kasperi N Korpinen; Emran Islam |
| Abstract: | We demonstrate how computer-based simulations could support cyber stress testing exercises through a three-step framework. First, cyber-attack scenarios are designed to target the systemic nodes of a payment network at different times, disrupting a major bank, critical service provider, large-value payment system, and a foreign exchange settlement system. Second, the stress resulting from the scenarios is simulated using transaction-level data, and its impact is measured through a range of risk metrics. And third, cyber preparedness is discussed to identify effective practices that could strengthen the cyber resilience of the financial sector. The exercise provides insights into the main vulnerabilities of the financial sector and key transmission channels under plausible scenarios that necessitate preemptive action and recovery and response measures. For example, simulation results for Finnish data suggest that end-of-day liquidity risk is most severe when a cyber-attack hits a major bank or several banks simultaneously through dependence on a common critical service provider, compared to an attack on a centralized payment system where effective queuing and liquidity-saving mechanisms can better support recovery. Outcomes could be aggravated under more severe and prolonged scenarios. |
| Keywords: | Cyber Resilience; Stress Testing; Simulation |
| Date: | 2025–05–02 |
| URL: | https://d.repec.org/n?u=RePEc:imf:imfwpa:2025/085 |
| By: | Syoiti Ninomiya; Yuji Shinozaki |
| Abstract: | This paper presents an algorithm for applying the high-order recombination method, originally introduced by Lyons and Litterer in ``High-order recombination and an application to cubature on Wiener space'' (Ann. Appl. Probab. 22(4):1301--1327, 2012), to practical problems in mathematical finance. A refined error analysis is provided, yielding a sharper condition for space partitioning. Based on this condition, a computationally feasible recursive partitioning algorithm is developed. Numerical examples are also included, demonstrating that the proposed algorithm effectively avoids the explosive growth in the cardinality of the support required to achieve high-order approximations. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.19717 |
| By: | Boulieris, Petros; Carballa-Smichowski, Bruno; Fourka, Maria Niki; Lianos, Ioannis |
| Abstract: | Following a period during which the two fields evolved separately, a consensus has emerged that competition and industrial policy are not inherently incompatible. This reflects broader intellectual shifts. Industrial policy is now viewed more favorably, not only for traditional development goals but also to strengthen technological capabilities for national security and secure global economic dominance. "Techno-nationalist'' approaches to industrial policy may conflict with global technology diffusion efforts addressing issues like climate change ("techno-globalism''). Despite recent developments in the intersection of competition and industrial policy, there is a lack of evidence on how techno-nationalist and techno-globalist approaches interact with competition policy goals. This article fills this gap by empirically assessing the competitive effects of policy measures. We use a text-as-data approach, combining AI-driven document analysis with structured classification criteria. The data show that techno-globalist industrial policies are generally more pro-competitive than techno-nationalist ones, due to their broader scope and ability to lower entry costs. Moreover, we find that certain policy instruments are primarily associated with anti-competitive criteria, while others tend to exhibit predominantly pro-competitive features. Our results provide a fine-grained characterization of new industrial policy design in light of competition policy goals. |
| Keywords: | Industrial policy, competition, techno-globalism, techno-nationalism, text-as-data, large language models, data analysis. |
| JEL: | D02 L0 L50 O25 |
| Date: | 2025–03 |
| URL: | https://d.repec.org/n?u=RePEc:pra:mprapa:124187 |
| By: | Gianandrea Lanzara; Matteo Santacesaria |
| Abstract: | Are there multiple equilibria in the spatial economy? This paper develops a unified framework that integrates systems of cities and regional models to address this question within a general geographic space. A key feature is the endogenous formation of commuting areas linking a continuum of residential locations to a finite set of potential business districts. Using tools from computational geometry and shape optimization, we derive sufficient conditions for the existence and uniqueness of spatial equilibria. For plausible parameter values, urban location is indeterminate, but, conditional on an urban system, city sizes are uniquely determined. The framework reconciles seemingly conflicting empirical findings on the role of geography and scale economies in shaping the spatial economy. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2504.21819 |