|
on Computational Economics |
| By: | Syoiti Ninomiya; Yuming Ma |
| Abstract: | A new deep-learning neural network architecture based on high-order weak approximation algorithms for stochastic differential equations (SDEs) is proposed. The architecture enables the efficient learning of martingales by deep learning models. The behaviour of deep neural networks based on this architecture, when applied to the problem of pricing financial derivatives, is also examined. The core of this new architecture lies in the high-order weak approximation algorithms of the explicit Runge--Kutta type, wherein the approximation is realised solely through iterative compositions and linear combinations of vector fields of the target SDEs. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.03789 |
| By: | Starobinsky, Mark |
| Abstract: | Large Language Models (LLMs) have propelled AI forward, yet they falter with static knowledge, unreliable outputs, and regulatory misalignment. Ontology-Enhanced AI, developed by OntoGuard AI, introduces a visionary framework that transcends these limits by weaving dynamic knowledge structures with sophisticated validation, tackling the Peak Data Problem head-on. Poised to transform enterprise AI with unparalleled adaptability and trust, this approach aligns with standards like GDPR and the EU AI Act. While proprietary breakthroughs remain under wraps due to a pending patent, this paper unveils the concept’s potential to captivate technical acquirers and licensees. |
| Date: | 2025–05–01 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:fh4ue_v1 |
| By: | Md. Yeasin Rahat; Rajan Das Gupta; Nur Raisa Rahman; Sudipto Roy Pritom; Samiur Rahman Shakir; Md Imrul Hasan Showmick; Md. Jakir Hossen |
| Abstract: | The prediction of foreign exchange rates, such as the US Dollar (USD) to Bangladeshi Taka (BDT), plays a pivotal role in global financial markets, influencing trade, investments, and economic stability. This study leverages historical USD/BDT exchange rate data from 2018 to 2023, sourced from Yahoo Finance, to develop advanced machine learning models for accurate forecasting. A Long Short-Term Memory (LSTM) neural network is employed, achieving an exceptional accuracy of 99.449%, a Root Mean Square Error (RMSE) of 0.9858, and a test loss of 0.8523, significantly outperforming traditional methods like ARIMA (RMSE 1.342). Additionally, a Gradient Boosting Classifier (GBC) is applied for directional prediction, with backtesting on a $10, 000 initial capital revealing a 40.82% profitable trade rate, though resulting in a net loss of $20, 653.25 over 49 trades. The study analyzes historical trends, showing a decline in BDT/USD rates from 0.012 to 0.009, and incorporates normalized daily returns to capture volatility. These findings highlight the potential of deep learning in forex forecasting, offering traders and policymakers robust tools to mitigate risks. Future work could integrate sentiment analysis and real-time economic indicators to further enhance model adaptability in volatile markets. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.09851 |
| By: | Dominik Stempie\'n; Janusz Gajda |
| Abstract: | We compare traditional approach of computing logarithmic returns with the fractional differencing method and its tempered extension as methods of data preparation before their usage in advanced machine learning models. Differencing parameters are estimated using multiple techniques. The empirical investigation is conducted on data from four major stock indices covering the most recent 10-year period. The set of explanatory variables is additionally extended with technical indicators. The effectiveness of the differencing methods is evaluated using both forecast error metrics and risk-adjusted return trading performance metrics. The findings suggest that fractional differentiation methods provide a suitable data transformation technique, improving the predictive model forecasting performance. Furthermore, the generated predictions appeared to be effective in constructing profitable trading strategies for both individual assets and a portfolio of stock indices. These results underline the importance of appropriate data transformation techniques in financial time series forecasting, supporting the application of memory-preserving techniques. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.19243 |
| By: | Pierre Brugi\`ere; Gabriel Turinici |
| Abstract: | Option pricing theory, such as the Black and Scholes (1973) model, provides an explicit solution to construct a strategy that perfectly hedges an option in a continuous-time setting. In practice, however, trading occurs in discrete time and often involves transaction costs, making the direct application of continuous-time solutions potentially suboptimal. Previous studies, such as those by Buehler et al. (2018), Buehler et al. (2019) and Cao et al. (2019), have shown that deep learning or reinforcement learning can be used to derive better hedging strategies than those based on continuous-time models. However, these approaches typically rely on a large number of trajectories (of the order of $10^5$ or $10^6$) to train the model. In this work, we show that using as few as 256 trajectories is sufficient to train a neural network that significantly outperforms, in the Geometric Brownian Motion framework, both the classical Black & Scholes formula and the Leland model, which is arguably one of the most effective explicit alternatives for incorporating transaction costs. The ability to train neural networks with such a small number of trajectories suggests the potential for more practical and simple implementation on real-time financial series. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.22836 |
| By: | Antonicelli, Margareth; Drago, Carlo; Costantiello, Alberto; Leogrande, Angelo |
| Abstract: | This study examines income inequality across Italian regions by integrating instrumental variable panel data models, k-means clustering, and machine learning algorithms. Using econometric techniques, we address endogeneity and identify causal relationships influencing regional disparities. K-means clustering, optimized with the elbow method, classifies Italian regions based on income inequality patterns, while machine-learning models, including random forest, support vector machines, and decision tree regression, predict inequality trends and key determinants. Informal employment, temporary employment, and overeducation also play a major role in influencing inequality. Clustering results confirm a permanent North-South economic divide and the most disadvantaged regions are Campania, Calabria, and Sicily. Among the machine learning models, the highest income disparities prediction accuracy comes with the use of Random Forest Regression. The findings emphasize the necessity of education-focused and digitally based policies and reforms of the labor market in an effort to enhance economic convergence. The study portrays the use of a combination of econometric and machine learning methods in the analysis of regional disparities and proposes a solid framework of policy-making with the intention of curbing economic disparities in Italy. |
| Date: | 2025–06–03 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:tk87m_v1 |
| By: | Dominik Stempie\'n; Robert \'Slepaczuk |
| Abstract: | This research systematically develops and evaluates various hybrid modeling approaches by combining traditional econometric models (ARIMA and ARFIMA models) with machine learning and deep learning techniques (SVM, XGBoost, and LSTM models) to forecast financial time series. The empirical analysis is based on two distinct financial assets: the S&P 500 index and Bitcoin. By incorporating over two decades of daily data for the S&P 500 and almost ten years of Bitcoin data, the study provides a comprehensive evaluation of forecasting methodologies across different market conditions and periods of financial distress. Models' training and hyperparameter tuning procedure is performed using a novel three-fold dynamic cross-validation method. The applicability of applied models is evaluated using both forecast error metrics and trading performance indicators. The obtained findings indicate that the proper construction process of hybrid models plays a crucial role in developing profitable trading strategies, outperforming their individual components and the benchmark Buy&Hold strategy. The most effective hybrid model architecture was achieved by combining the econometric ARIMA model with either SVM or LSTM, under the assumption of a non-additive relationship between the linear and nonlinear components. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.19617 |
| By: | Hariom Tatsat (Barclays); Ariye Shater (Barclays) |
| Abstract: | Large Language Models (LLMs) exhibit remarkable capabilities across a spectrum of tasks in financial services, including report generation, chatbots, sentiment analysis, regulatory compliance, investment advisory, financial knowledge retrieval, and summarization. However, their intrinsic complexity and lack of transparency pose significant challenges, especially in the highly regulated financial sector, where interpretability, fairness, and accountability are critical. As far as we are aware, this paper presents the first application in the finance domain of understanding and utilizing the inner workings of LLMs through mechanistic interpretability, addressing the pressing need for transparency and control in AI systems. Mechanistic interpretability is the most intuitive and transparent way to understand LLM behavior by reverse-engineering their internal workings. By dissecting the activations and circuits within these models, it provides insights into how specific features or components influence predictions - making it possible not only to observe but also to modify model behavior. In this paper, we explore the theoretical aspects of mechanistic interpretability and demonstrate its practical relevance through a range of financial use cases and experiments, including applications in trading strategies, sentiment analysis, bias, and hallucination detection. While not yet widely adopted, mechanistic interpretability is expected to become increasingly vital as adoption of LLMs increases. Advanced interpretability tools can ensure AI systems remain ethical, transparent, and aligned with evolving financial regulations. In this paper, we have put special emphasis on how these techniques can help unlock interpretability requirements for regulatory and compliance purposes - addressing both current needs and anticipating future expectations from financial regulators globally. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.24650 |
| By: | Francesco Morri; H\'el\`ene Le Cadre; Pierre Gruet; Luce Brotcorne |
| Abstract: | We characterize zonal ancillary market coupling relying on noncooperative game theory. To that purpose, we formulate the ancillary market as a multi-leader single follower bilevel problem, that we subsequently cast as a generalized Nash game with side constraints and nonconvex feasibility sets. We determine conditions for equilibrium existence and show that the game has a generalized potential game structure. To compute market equilibrium, we rely on two exact approaches: an integrated optimization approach and Gauss-Seidel best-response, that we compare against multi-agent deep reinforcement learning. On real data from Germany and Austria, simulations indicate that multi-agent deep reinforcement learning achieves the smallest convergence rate but requires pretraining, while best-response is the slowest. On the economics side, multi-agent deep reinforcement learning results in smaller market costs compared to the exact methods, but at the cost of higher variability in the profit allocation among stakeholders. Further, stronger coupling between zones tends to reduce costs for larger zones. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.03288 |
| By: | Jiaxiang Chen; Mingxi Zou; Zhuo Wang; Qifan Wang; Dongning Sun; Chi Zhang; Zenglin Xu |
| Abstract: | Financial decision-making presents unique challenges for language models, demanding temporal reasoning, adaptive risk assessment, and responsiveness to dynamic events. While large language models (LLMs) show strong general reasoning capabilities, they often fail to capture behavioral patterns central to human financial decisions-such as expert reliance under information asymmetry, loss-averse sensitivity, and feedback-driven temporal adjustment. We propose FinHEAR, a multi-agent framework for Human Expertise and Adaptive Risk-aware reasoning. FinHEAR orchestrates specialized LLM-based agents to analyze historical trends, interpret current events, and retrieve expert-informed precedents within an event-centric pipeline. Grounded in behavioral economics, it incorporates expert-guided retrieval, confidence-adjusted position sizing, and outcome-based refinement to enhance interpretability and robustness. Empirical results on curated financial datasets show that FinHEAR consistently outperforms strong baselines across trend prediction and trading tasks, achieving higher accuracy and better risk-adjusted returns. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.09080 |
| By: | Lijie Ding; Egang Lu; Kin Cheung |
| Abstract: | We introduce a fast and flexible Machine Learning (ML) framework for pricing derivative products whose valuation depends on volatility surfaces. By parameterizing volatility surfaces with the 5-parameter stochastic volatility inspired (SVI) model augmented by a one-factor term structure adjustment, we first generate numerous volatility surfaces over realistic ranges for these parameters. From these synthetic market scenarios, we then compute high-accuracy valuations using conventional methodologies for two representative products: the fair strike of a variance swap and the price and Greeks of an American put. We then train the Gaussian Process Regressor (GPR) to learn the nonlinear mapping from the input risk factors, which are the volatility surface parameters, strike and interest rate, to the valuation outputs. Once trained, We use the GPR to perform out-of-sample valuations and compare the results against valuations using conventional methodologies. Our ML model achieves very accurate results of $0.5\%$ relative error for the fair strike of variance swap and $1.7\% \sim 3.5\%$ relative error for American put prices and first-order Greeks. More importantly, after training, the model computes valuations almost instantly, yielding a three to four orders of magnitude speedup over Crank-Nicolson finite-difference method for American puts, enabling real-time risk analytics, dynamic hedging and large-scale scenario analysis. Our approach is general and can be extended to other path-dependent derivative products with early-exercise features, paving the way for hybrid quantitative engines for modern financial systems. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.22957 |
| By: | Zikun Ye; Hema Yoganarasimhan |
| Abstract: | Large Language Models (LLMs) are increasingly used in systems that retrieve and summarize content from multiple sources, such as search engines and AI assistants. While these models enhance user experience by generating coherent summaries, they obscure the contributions of original content creators, raising concerns about credit attribution and compensation. We address the challenge of valuing individual documents used in LLM-generated summaries. We propose using Shapley values, a game-theoretic method that allocates credit based on each document's marginal contribution. Although theoretically appealing, Shapley values are expensive to compute at scale. We therefore propose Cluster Shapley, an efficient approximation algorithm that leverages semantic similarity between documents. By clustering documents using LLM-based embeddings and computing Shapley values at the cluster level, our method significantly reduces computation while maintaining attribution quality. We demonstrate our approach to a summarization task using Amazon product reviews. Cluster Shapley significantly reduces computational complexity while maintaining high accuracy, outperforming baseline methods such as Monte Carlo sampling and Kernel SHAP with a better efficient frontier. Our approach is agnostic to the exact LLM used, the summarization process used, and the evaluation procedure, which makes it broadly applicable to a variety of summarization settings. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.23842 |
| By: | Maximilian Andres (University of Bremen); Lisa Bruttel (University of Potsdam, CEPA) |
| Abstract: | While an influential body of economic literature shows that allowing for communication between firms increases collusion in oligopolies, so far we have only anecdotal evidence on the precise communication content that helps firms to coordinate their behavior. In this paper, we conduct a primary-data meta-study on oligopoly experiments and use a machine learning approach to identify systematic patterns in the communication content across studies. Starting with the communication topics mentioned most often in the literature (agreements, joint benefit, threat of punishment, promise/trust), we use a semi-supervised approach to detect all relevant topics. In a second step, we study the effect of these topics on the rate of collusion among the firms. We find that agreements on specific behavior are decisive for the strong positive effect of communication on collusion, while other communication topics have no effect. |
| Keywords: | collusion, communication, machine learning, meta-study, experiment |
| JEL: | C92 D43 L41 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:pot:cepadp:88 |
| By: | Jasper Rou |
| Abstract: | Option pricing often requires solving partial differential equations (PDEs). Although deep learning-based PDE solvers have recently emerged as quick solutions to this problem, their empirical and quantitative accuracy remain not well understood, hindering their real-world applicability. In this research, our aim is to offer actionable insights into the utility of deep PDE solvers for practical option pricing implementation. Through comparative experiments in both the Black--Scholes and the Heston model, we assess the empirical performance of two neural network algorithms to solve PDEs: the Deep Galerkin Method and the Time Deep Gradient Flow method (TDGF). We determine their empirical convergence rates and training time as functions of (i) the number of sampling stages, (ii) the number of samples, (iii) the number of layers, and (iv) the number of nodes per layer. For the TDGF, we also consider the order of the discretization scheme and the number of time steps. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.05121 |
| By: | Chung I Lu; Julian Sester; Aijia Zhang |
| Abstract: | We propose a novel distributionally robust $Q$-learning algorithm for the non-tabular case accounting for continuous state spaces where the state transition of the underlying Markov decision process is subject to model uncertainty. The uncertainty is taken into account by considering the worst-case transition from a ball around a reference probability measure. To determine the optimal policy under the worst-case state transition, we solve the associated non-linear Bellman equation by dualising and regularising the Bellman operator with the Sinkhorn distance, which is then parameterized with deep neural networks. This approach allows us to modify the Deep Q-Network algorithm to optimise for the worst case state transition. We illustrate the tractability and effectiveness of our approach through several applications, including a portfolio optimisation task based on S\&{P}~500 data. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.19058 |
| By: | Erfan Loghmani |
| Abstract: | Large language models are being widely used across industries to generate content that contributes directly to key performance metrics, such as conversion rates. Pretrained models, however, often fall short when it comes to aligning with human preferences or optimizing for business objectives. As a result, fine-tuning with good-quality labeled data is essential to guide models to generate content that achieves better results. Controlled experiments, like A/B tests, can provide such data, but they are often expensive and come with significant engineering and logistical challenges. Meanwhile, companies have access to a vast amount of historical (observational) data that remains underutilized. In this work, we study the challenges and opportunities of fine-tuning LLMs using observational data. We show that while observational outcomes can provide valuable supervision, directly fine-tuning models on such data can lead them to learn spurious correlations. We present empirical evidence of this issue using various real-world datasets and propose DeconfoundLM, a method that explicitly removes the effect of known confounders from reward signals. Using simulation experiments, we demonstrate that DeconfoundLM improves the recovery of causal relationships and mitigates failure modes found in fine-tuning methods that ignore or naively incorporate confounding variables. Our findings highlight that while observational data presents risks, with the right causal corrections, it can be a powerful source of signal for LLM alignment. Please refer to the project page for code and related resources. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.00152 |
| By: | Jiahao Yang; Ran Fang; Ming Zhang; Jun Zhou |
| Abstract: | In high-frequency trading (HFT), leveraging limit order books (LOB) to model stock price movements is crucial for achieving profitable outcomes. However, this task is challenging due to the high-dimensional and volatile nature of the original data. Even recent deep learning models often struggle to capture price movement patterns effectively, particularly without well-designed features. We observed that raw LOB data exhibits inherent symmetry between the ask and bid sides, and the bid-ask differences demonstrate greater stability and lower complexity compared to the original data. Building on this insight, we propose a novel approach in which leverages the Siamese architecture to enhance the performance of existing deep learning models. The core idea involves processing the ask and bid sides separately using the same module with shared parameters. We applied our Siamese-based methods to several widely used strong baselines and validated their effectiveness using data from 14 military industry stocks in the Chinese A-share market. Furthermore, we integrated multi-head attention (MHA) mechanisms with the Long Short-Term Memory (LSTM) module to investigate its role in modeling stock price movements. Our experiments used raw data and widely used Order Flow Imbalance (OFI) features as input with some strong baseline models. The results show that our method improves the performance of strong baselines in over 75$% of cases, excluding the Multi-Layer Perception (MLP) baseline, which performed poorly and is not considered practical. Furthermore, we found that Multi-Head Attention can enhance model performance, particularly over shorter forecasting horizons. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.22678 |
| By: | Shanyu Han; Yang Liu; Xiang Yu |
| Abstract: | We propose a reinforcement learning (RL) framework under a broad class of risk objectives, characterized by convex scoring functions. This class covers many common risk measures, such as variance, Expected Shortfall, entropic Value-at-Risk, and mean-risk utility. To resolve the time-inconsistency issue, we consider an augmented state space and an auxiliary variable and recast the problem as a two-state optimization problem. We propose a customized Actor-Critic algorithm and establish some theoretical approximation guarantees. A key theoretical contribution is that our results do not require the Markov decision process to be continuous. Additionally, we propose an auxiliary variable sampling method inspired by the alternating minimization algorithm, which is convergent under certain conditions. We validate our approach in simulation experiments with a financial application in statistical arbitrage trading, demonstrating the effectiveness of the algorithm. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.04553 |
| By: | Geyue Sun; Xiao Liu; Tomas Williams; Roberto Samaniego |
| Abstract: | We construct a novel event-level Capital Control Measures (CCM) dataset covering 196 countries from 1999 to 2023 by leveraging prompt-based large language models (LLMs). The dataset enables event study analysis and cross-country comparisons based on rich policy attributes, including action type, intensity, direction, implementing entity, and other multidimensional characteristics. Using a two-step prompt framework with GPT-4.1, we extract structured information from the IMF's Annual Report on Exchange Arrangements and Exchange Restrictions (AREAER), resulting in 5, 198 capital control events with 27 annotated fields and corresponding model reasoning. Secondly, to facilitate real-time classification and extension to external sources, we fine-tune an open-source Meta Llama 3.1-8B model, named CCM-Llama, trained on AREAER change logs and final status reports. The model achieves 90.09\% accuracy in category classification and 99.55\% in status prediction. Finally, we apply the CCM dataset in an empirical application: an event study on China, Australia, and the US. The results show that inward capital control measures significantly reduce fund inflows within one month, and restrictive policies tend to have stronger effects than liberalizing ones, with notable heterogeneity across countries. Our work contributes to the growing literature on the use of LLMs in economics by providing both a novel high-frequency policy dataset and a replicable framework for automated classification of capital control events from diverse and evolving information sources. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.23025 |
| By: | Olivier Toubia; George Z. Gui; Tianyi Peng; Daniel J. Merlau; Ang Li; Haozhe Chen |
| Abstract: | LLM-based digital twin simulation, where large language models are used to emulate individual human behavior, holds great promise for research in AI, social science, and digital experimentation. However, progress in this area has been hindered by the scarcity of real, individual-level datasets that are both large and publicly available. This lack of high-quality ground truth limits both the development and validation of digital twin methodologies. To address this gap, we introduce a large-scale, public dataset designed to capture a rich and holistic view of individual human behavior. We survey a representative sample of $N = 2, 058$ participants (average 2.42 hours per person) in the US across four waves with 500 questions in total, covering a comprehensive battery of demographic, psychological, economic, personality, and cognitive measures, as well as replications of behavioral economics experiments and a pricing survey. The final wave repeats tasks from earlier waves to establish a test-retest accuracy baseline. Initial analyses suggest the data are of high quality and show promise for constructing digital twins that predict human behavior well at the individual and aggregate levels. By making the full dataset publicly available, we aim to establish a valuable testbed for the development and benchmarking of LLM-based persona simulations. Beyond LLM applications, due to its unique breadth and scale the dataset also enables broad social science research, including studies of cross-construct correlations and heterogeneous treatment effects. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.17479 |
| By: | Schraml, Christopher |
| Abstract: | The exponential growth of online video content presents significant opportunities for marketing researchers to analyze online consumer behavior. However, effectively using this data requires advanced methods for managing the complexity and sheer volume of video data. This research presents a systematic literature review of video analytics in marketing research, outlining the types of information extracted automatically from video data and examining the specific methodologies employed by researchers. Furthermore, we evaluate a multimodal large language model, specifically ChatGPT-4o, for complex, zero-shot image coding tasks in marketing research. Building on these image coding capabilities, we introduce a novel multimodal large language model-based pipeline for zero-shot video analysis. This innovative method allows researchers to efficiently extract meaningful information from video data without prior training or labeled datasets. Our findings highlight how multimodal large language models can advance video analytics into a scalable and cost-effective tool for marketing research. |
| Date: | 2025–05–21 |
| URL: | https://d.repec.org/n?u=RePEc:osf:osfxxx:63nbc_v1 |
| By: | Joshua S. Gans |
| Abstract: | This paper analyzes the design of tests to distinguish human from artificial intelligence through the lens of information design. We identify a fundamental asymmetry: while AI systems can strategically underperform to mimic human limitations, they cannot overperform beyond their capabilities. This leads to our main contribution—the concept of capability inversion domains, where AIs fail detection not through inferior performance, but by performing “suspiciously well” when they overestimate human capabilities. We show that if an AI significantly overestimates human ability in even one domain, it cannot reliably pass an optimally designed test. This insight reverses conventional intuition: effective tests should target not what humans do well, but the specific patterns of human imperfection that AIs systematically misunderstand. We identify structural sources of persistent misperception—including the difficulty of learning about failure from successful examples and fundamental differences in embodied experience—that make certain capability inversions exploitable for detection even as AI systems improve. |
| JEL: | C72 D82 D83 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:33893 |
| By: | Perin, Augusto O.; Castro, Hélder U.; Filho, Euclério B. O.; Martins, Joberto S. B. Prof. Dr. (Salvador University - UNIFACS) |
| Abstract: | The smart city strategy drives the creation of innovative solutions to the challenges of urban planning in cities. With its data collection capacity, the Internet of Things (IoT) is a key component of Information and Communication Technologies in smart cities. Likewise, Artificial Intelligence (AI) provides a set of vital tools for data analysis and service optimization. This article presents a discussion on the use of IoT in conjunction with Artificial Intelligence to obtain new solutions for planning, management, and services in smart cities. The analysis is based on a literature review that seeks to identify answers to questions involving urban planning in cities, the structuring themes of smart cities, AI techniques, and data collection. The results point to a powerful synergy between IoT and AI, aiming to develop innovative, effective, and integrated services to address the challenges of contemporary urban planning. |
| Date: | 2024–12–17 |
| URL: | https://d.repec.org/n?u=RePEc:osf:socarx:r36yg_v1 |
| By: | Lucien Chaffa; Martin Trépanier; Thierry Warin |
| Abstract: | This study investigates the potential of machine learning (ML) methods to enhance the estimation of the gravity model, a cornerstone of international trade analysis that explains trade flows based on economic size and distance. Traditionally estimated using methods such as the Poisson Pseudo Maximum Likelihood (PPML) approach, gravity models often struggle to fully capture nonlinear relationships and intricate interactions among variables. Leveraging data from Canada and the US, one of the largest bilateral trading relationships in the world, this paper conducts a comparative analysis of traditional and ML approaches. The findings reveal that ML methods can significantly outperform traditional approaches in predicting trade flows, offering a robust alternative for capturing the complexities of global trade dynamics. These results underscore the value of integrating ML techniques into trade policy analysis, providing policymakers and economists with improved tools for decision-making. Cette étude examine le potentiel des méthodes d'apprentissage automatique (ML) pour améliorer l'estimation du modèle de gravité, une méthode clé de l'analyse du commerce international qui explique les flux commerciaux en fonction de la taille de l'économie et de la distance. Traditionnellement estimés à l'aide de méthodes telles que l'approche du pseudo-maximum de vraisemblance de Poisson (PPML), les modèles de gravité ont souvent du mal à saisir pleinement les relations non linéaires et les interactions complexes entre les variables. En s'appuyant sur les données du Canada et des États-Unis, l'une des relations commerciales bilatérales les plus importantes au monde, cet article effectue une analyse comparative des approches traditionnelles et des approches par apprentissage automatique. Les résultats révèlent que les méthodes de ML peuvent être nettement plus performantes que les approches traditionnelles pour prédire les flux commerciaux, offrant ainsi une alternative robuste pour saisir les complexités de la dynamique du commerce mondial. Ces résultats soulignent la valeur de l'intégration des techniques de ML dans l'analyse de la politique commerciale, fournissant aux décideurs politiques et aux économistes des outils améliorés pour la prise de décision. |
| Keywords: | Gravity Model, PPML Machine Learning, International Trade, Trade Policy Analysis, Modèle de gravité, PPML, apprentissage automatique, commerce international, analyse de la politique commerciale |
| JEL: | F10 F14 C13 C45 |
| Date: | 2025–05–20 |
| URL: | https://d.repec.org/n?u=RePEc:cir:cirwor:2025s-14 |
| By: | Jianhao Lin; Lexuan Sun; Yixin Yan |
| Abstract: | We introduce a novel framework for simulating macroeconomic expectation formation using Large Language Model-Empowered Agents (LLM Agents). By constructing thousands of LLM Agents equipped with modules for personal characteristics, prior expectations, and knowledge, we replicate a survey experiment involving households and experts on inflation and unemployment. Our results show that although the expectations and thoughts generated by LLM Agents are more homogeneous than those of human participants, they still effectively capture key heterogeneity across agents and the underlying drivers of expectation formation. Furthermore, a module-ablation exercise highlights the critical role of prior expectations in simulating such heterogeneity. This approach complements traditional survey methods and offers new insights into AI behavioral science in macroeconomic research. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.17648 |
| By: | Daniel F. Villarraga; Ricardo A. Daziano |
| Abstract: | Discrete choice models (DCMs) are used to analyze individual decision-making in contexts such as transportation choices, political elections, and consumer preferences. DCMs play a central role in applied econometrics by enabling inference on key economic variables, such as marginal rates of substitution, rather than focusing solely on predicting choices on new unlabeled data. However, while traditional DCMs offer high interpretability and support for point and interval estimation of economic quantities, these models often underperform in predictive tasks compared to deep learning (DL) models. Despite their predictive advantages, DL models remain largely underutilized in discrete choice due to concerns about their lack of interpretability, unstable parameter estimates, and the absence of established methods for uncertainty quantification. Here, we introduce a deep learning model architecture specifically designed to integrate with approximate Bayesian inference methods, such as Stochastic Gradient Langevin Dynamics (SGLD). Our proposed model collapses to behaviorally informed hypotheses when data is limited, mitigating overfitting and instability in underspecified settings while retaining the flexibility to capture complex nonlinear relationships when sufficient data is available. We demonstrate our approach using SGLD through a Monte Carlo simulation study, evaluating both predictive metrics--such as out-of-sample balanced accuracy--and inferential metrics--such as empirical coverage for marginal rates of substitution interval estimates. Additionally, we present results from two empirical case studies: one using revealed mode choice data in NYC, and the other based on the widely used Swiss train choice stated preference data. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.18077 |
| By: | Jean-Paul Daemen; Silvia Leoni |
| Abstract: | This paper employs agent-based modelling to explore the factors driving the high rate of tertiary education completion in the Netherlands. We examine the interplay of economic motivations, such as expected wages and financial constraints, alongside sociological and psychological influences, including peer effects, student disposition, personality, and geographic accessibility. Through simulations, we analyse the sustainability of these trends and evaluate the impact of educational policies, such as student grants and loans, on enrollment and borrowing behaviour among students from different socioeconomic backgrounds, further considering implications for the Dutch labour market. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.01142 |
| By: | Kristoffer Andersson; Alessandro Gnoatto; Camilo Andr\'es Garc\'ia Trillos |
| Abstract: | We present the first deep-learning solver for backward stochastic Volterra integral equations (BSVIEs) and their fully-coupled forward-backward variants. The method trains a neural network to approximate the two solution fields in a single stage, avoiding the use of nested time-stepping cycles that limit classical algorithms. For the decoupled case we prove a non-asymptotic error bound composed of an a posteriori residual plus the familiar square root dependence on the time step. Numerical experiments confirm this rate and reveal two key properties: \emph{scalability}, in the sense that accuracy remains stable from low dimension up to 500 spatial variables while GPU batching keeps wall-clock time nearly constant; and \emph{generality}, since the same method handles coupled systems whose forward dynamics depend on the backward solution. These results open practical access to a family of high-dimensional, path-dependent problems in stochastic control and quantitative finance. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.18297 |
| By: | Shanyan Lai |
| Abstract: | In this study, MLP models with dynamic structure are applied to factor models for asset pricing tasks. Concretely, the MLP pyramid model structure was employed on firm-characteristic-sorted portfolio factors for modelling the large-capital US stocks. It was further developed as a practicable factor investing strategy based on the predictions. The main findings in this chapter were evaluated from two angles: model performance and investing performance, which were compared from the periods with and without COVID-19. The empirical results indicated that with the restrictions of the data size, the MLP models no longer perform "deeper, better", while the proposed MLP models with two and three hidden layers have higher flexibility to model the factors in this case. This study also verified the idea of previous works that MLP models for factor investing have more meaning in the downside risk control than in pursuing the absolute annual returns. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.01921 |
| By: | Alberto Cabezas; Carlotta Montorsi |
| Abstract: | Social sciences rely on counterfactual analysis using surveys and administrative data, generally depending on strong assumptions or the existence of suitable control groups, to evaluate policy interventions and estimate causal effects. We propose a novel approach that leverages the Transformer architecture to simulate counterfactual life trajectories from large-scale administrative records. Our contributions are: the design of a novel encoding method that transforms longitudinal administrative data to sequences and the proposal of a generative model tailored to life sequences with overlapping events across life domains. We test our method using data from the Istituto Nazionale di Previdenza Sociale (INPS), showing that it enables the realistic and coherent generation of life trajectories. This framework offers a scalable alternative to classical counterfactual identification strategy, such as difference-in-differences and synthetic controls, particularly in contexts where these methods are infeasible or their assumptions unverifiable. We validate the model's utility by comparing generated life trajectories against established findings from causal studies, demonstrating its potential to enrich labour market research and policy evaluation through individual-level simulations. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.01874 |
| By: | Stephen J. Lee; Cailinn Drouin |
| Abstract: | We present a novel framework for high-resolution forecasting of residential heating and electricity demand using probabilistic deep learning models. We focus specifically on providing hourly building-level electricity and heating demand forecasts for the residential sector. Leveraging multimodal building-level information -- including data on building footprint areas, heights, nearby building density, nearby building size, land use patterns, and high-resolution weather data -- and probabilistic modeling, our methods provide granular insights into demand heterogeneity. Validation at the building level underscores a step change improvement in performance relative to NREL's ResStock model, which has emerged as a research community standard for residential heating and electricity demand characterization. In building-level heating and electricity estimation backtests, our probabilistic models respectively achieve RMSE scores 18.3\% and 35.1\% lower than those based on ResStock. By offering an open-source, scalable, high-resolution platform for demand estimation and forecasting, this research advances the tools available for policymakers and grid planners, contributing to the broader effort to decarbonize the U.S. building stock and meeting climate objectives. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.22873 |
| By: | Pascal Michaillat |
| Abstract: | This paper develops a new method for detecting US recessions in real time. The method constructs millions of recession classifiers by combining unemployment and vacancy data to reduce detection noise. Classifiers are then selected to avoid both false negatives (missed recessions) and false positives (nonexistent recessions). The selected classifiers are therefore perfect, in that they identify all 15 historical recessions in the training period without any false positives. By further selecting classifiers that lie on the high-precision segment of the anticipation-precision frontier, the method optimizes early detection without sacrificing precision. On average, over 1929--2021, the classifier ensemble signals recessions 2.2 months after their true onset, with a standard deviation of 1.9 months. Applied to May 2025 data, the classifier ensemble gives a 71% probability that the US economy is currently in recession. Backtesting to 2004, 1984, and 1964 confirms the algorithm's reliability. Algorithms trained on limited historical windows continue to detect all subsequent recessions without errors. Furthermore, they all detect the Great Recession by mid-2008 -- even when they are only trained on data up to 1984 or 1964. The classifier ensembles trained on 1929--2004, 1929--1984, and 1929--1964 data give a current recession probability of 58%, 83%, and 25%, respectively. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.09664 |