|
on Computational Economics |
| By: | M.Jahangir Alam; Shane Boyle; Huiyu Li; Tatevik Sekhposyan |
| Abstract: | Recent research suggests that generic large language models (LLMs) can match the accuracy of traditional methods when forecasting macroeconomic variables in pseudo out-of-sample settings generated via prompts. This paper assesses the out-of-sample forecasting accuracy of LLMs by eliciting real-time forecasts of U.S. inflation from ChatGPT. We find that out-of-sample predictions are largely inaccurate and stale, even though forecasts generated in pseudo out-of-sample environments are comparable to existing benchmarks. Our results underscore the importance of out-of-sample benchmarking for LLM predictions. |
| Keywords: | large language models; generative AI; inflation forecasting |
| JEL: | C45 E31 E37 |
| Date: | 2026–02–05 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedfwp:102407 |
| By: | Zheng Li |
| Abstract: | This study addresses the low-volatility Chinese Public Real Estate Investment Trusts (REITs) market, proposing a large language model (LLM)-driven trading framework based on multi-agent collaboration. The system constructs four types of analytical agents-announcement, event, price momentum, and market-each conducting analysis from different dimensions; then the prediction agent integrates these multi-source signals to output directional probability distributions across multiple time horizons, then the decision agent generates discrete position adjustment signals based on the prediction results and risk control constraints, thereby forming a closed loop of analysis-prediction-decision-execution. This study further compares two prediction model pathways: for the prediction agent, directly calling the general-purpose large model DeepSeek-R1 versus using a specialized small model Qwen3-8B fine-tuned via supervised fine-tuning and reinforcement learning alignment. In the backtest from October 2024 to October 2025, both agent-based strategies significantly outperformed the buy-and-hold benchmark in terms of cumulative return, Sharpe ratio, and maximum drawdown. The results indicate that the multi-agent framework can effectively enhance the risk-adjusted return of REITs trading, and the fine-tuned small model performs close to or even better than the general-purpose large model in some scenarios. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00082 |
| By: | Felipe A. Csaszar; Aticus Peterson; Daniel Wilde |
| Abstract: | Can artificial intelligence outperform humans at strategic foresight -- the capacity to form accurate judgments about uncertain, high-stakes outcomes before they unfold? We address this question through a fully prospective prediction tournament using live Kickstarter crowdfunding projects. Thirty U.S.-based technology ventures, launched after the training cutoffs of all models studied, were evaluated while fundraising remained in progress and outcomes were unknown. A diverse suite of frontier and open-weight large language models (LLMs) completed 870 pairwise comparisons, producing complete rankings of predicted fundraising success. We benchmarked these forecasts against 346 experienced managers recruited via Prolific and three MBA-trained investors working under monitored conditions. The results are striking: human evaluators achieved rank correlations with actual outcomes between 0.04 and 0.45, while several frontier LLMs exceeded 0.60, with the best (Gemini 2.5 Pro) reaching 0.74 -- correctly ordering nearly four of every five venture pairs. These differences persist across multiple performance metrics and robustness checks. Neither wisdom-of-the-crowd ensembles nor human-AI hybrid teams outperformed the best standalone model. |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.01684 |
| By: | Brandon Yee; Krishna Sharma |
| Abstract: | Behavioral parameters such as loss aversion, herding, and extrapolation are central to asset pricing models but remain difficult to measure reliably. We develop a framework that treats large language models (LLMs) as calibrated measurement instruments for behavioral parameters. Using four models and 24{, }000 agent--scenario pairs, we document systematic rationality bias in baseline LLM behavior, including attenuated loss aversion, weak herding, and near-zero disposition effects relative to human benchmarks. Profile-based calibration induces large, stable, and theoretically coherent shifts in several parameters, with calibrated loss aversion, herding, extrapolation, and anchoring reaching or exceeding benchmark magnitudes. To assess external validity, we embed calibrated parameters in an agent-based asset pricing model, where calibrated extrapolation generates short-horizon momentum and long-horizon reversal patterns consistent with empirical evidence. Our results establish measurement ranges, calibration functions, and explicit boundaries for eight canonical behavioral biases. |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.01022 |
| By: | Keywan Christian Rasekhschaffe |
| Abstract: | We study whether generative AI can automate feature discovery in U.S. equities. Using large language models with retrieval-augmented generation and structured/programmatic prompting, we synthesize economically motivated features from analyst, options, and price-volume data. These features are then used as inputs to a tabular machine-learning model to forecast short-horizon returns. Across multiple datasets, AI-generated features are consistently competitive with baselines, with Sharpe improvements ranging from 14% to 91% depending on dataset and configuration. Retrieval quality is pivotal: better knowledge bases materially improve outcomes. The AI-generated signals are weakly correlated with traditional features, supporting combination. Overall, generative AI can meaningfully augment feature discovery when retrieval quality is controlled, producing interpretable signals while reducing manual engineering effort. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00196 |
| By: | Oskar V{\aa}le; Shiliang Zhang; Sabita Maharjan; Gro Kl{\ae}boe |
| Abstract: | The balancing market in the energy sector plays a critical role in physically and financially balancing the supply and demand. Modeling dynamics in the balancing market can provide valuable insights and prognosis for power grid stability and secure energy supply. While complex machine learning models can achieve high accuracy, their black-box nature severely limits the model interpretability. In this paper, we explore the trade-off between model accuracy and interpretability for the energy balancing market. Particularly, we take the example of forecasting manual frequency restoration reserve (mFRR) activation price in the balancing market using real market data from different energy price zones. We explore the interpretability of mFRR forecasting using two models: extreme gradient boosting (XGBoost) machine and explainable boosting machine (EBM). We also integrate the two models, and we benchmark all the models against a baseline naive model. Our results show that EBM provides forecasting accuracy comparable to XGBoost while yielding a considerable level of interpretability. Our analysis also underscores the challenge of accurately predicting the mFRR price for the instances when the activation price deviates significantly from the spot price. Importantly, EBM's interpretability features reveal insights into non-linear mFRR price drivers and regional market dynamics. Our study demonstrates that EBM is a viable and valuable interpretable alternative to complex black-box AI models in the forecast for the balancing market. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00049 |
| By: | Eurydice Fotopoulou; Iyke Maduako; M. Belen Sbrancia; Prachi Srivastava |
| Abstract: | The absence of reliable data on fundamental economic indicators (e.g. real GDP), combined with structural shifts in the economy, can severely constrain the ability to conduct accurate macroeconomic analysis and forecasting. This paper explores alternatives to address data limitations by integrating machine learning and satellite data to estimate real GDP. Specifically, it finds that incorporating satellite-based nightlight data into a random forest model significantly improves the accuracy of quarterly GDP growth estimates compared with models relying solely on traditional indicators. This empirical application contributes to the emerging nowcasting field to enhance economic forecasting in economies with significant data gaps. |
| Keywords: | Macroeconomic forecast; Machine learning; Nowcasting; GDP; Satellite data; Random Forest |
| Date: | 2026–01–30 |
| URL: | https://d.repec.org/n?u=RePEc:imf:imfwpa:2026/020 |
| By: | Walid Siala (SnT, University of Luxembourg, Luxembourg); Ahmed Khanfir (RIADI, ENSI, University of Manouba, Tunisia; SnT, University of Luxembourg, Luxembourg); Mike Papadakis (SnT, University of Luxembourg, Luxembourg) |
| Abstract: | This paper addresses stock price movement prediction by leveraging LLM-based news sentiment analysis. Earlier works have largely focused on proposing and assessing sentiment analysis models and stock movement prediction methods, however, separately. Although promising results have been achieved, a clear and in-depth understanding of the benefit of the news sentiment to this task, as well as a comprehensive assessment of different architecture types in this context, is still lacking. Herein, we conduct an evaluation study that compares 3 different LLMs, namely, DeBERTa, RoBERTa and FinBERT, for sentiment-driven stock prediction. Our results suggest that DeBERTa outperforms the other two models with an accuracy of 75% and that an ensemble model that combines the three models can increase the accuracy to about 80%. Also, we see that sentiment news features can benefit (slightly) some stock market prediction models, i.e., LSTM-, PatchTST- and tPatchGNN-based classifiers and PatchTST- and TimesNet-based regression tasks models. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00086 |
| By: | Yuanhong Wu; Wei Ye; Jingyan Xu; D. Frank Hsu |
| Abstract: | In this work, we propose to apply a new model fusion and learning paradigm, known as Combinatorial Fusion Analysis (CFA), to the field of Bitcoin price prediction. Price prediction of financial product has always been a big topic in finance, as the successful prediction of the price can yield significant profit. Every machine learning model has its own strength and weakness, which hinders progress toward robustness. CFA has been used to enhance models by leveraging rank-score characteristic (RSC) function and cognitive diversity in the combination of a moderate set of diverse and relatively well-performed models. Our method utilizes both score and rank combinations as well as other weighted combination techniques. Key metrics such as RMSE and MAPE are used to evaluate our methodology performance. Our proposal presents a notable MAPE performance of 0.19\%. The proposed method greatly improves upon individual model performance, as well as outperforms other Bitcoin price prediction models. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00037 |
| By: | Federico Cacciamani; Roberto Daluiso; Marco Pinciroli; Michele Trapletti; Edoardo Vittori |
| Abstract: | In finance, sequential decision problems are often faced, for which reinforcement learning (RL) emerges as a promising tool for optimisation without the need of analytical tractability. However, the objective of classical RL is the expected cumulated reward, while financial applications typically require a trade-off between return and risk. In this work, we focus on settings where one cares about the time split of the total return, ruling out most risk-aware generalisations of RL which optimise a risk measure defined on the latter. We notice that a preference for homogeneous splits, which we found satisfactory for hedging, can be unfit for other problems, and therefore propose a new risk metric which still penalises uncertainty of the single rewards, but allows for an arbitrary planning of their target levels. We study the properties of the resulting objective and the generalisation of learning algorithms to optimise it. Finally, we show numerical results on toy examples. |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.12030 |
| By: | Anauati, María Victoria; Romero, María Noelia; Baraldi, Lucia; Sosa Escudero, Walter; Tommasi, Mariano |
| Abstract: | Recidivism is a persistent challenge for criminal justice systems worldwide, yet evidence from Latin America remains scarce. This study addresses that gap through three contributions. First, it reviews the individual, institutional, and contextual determinants of recidivism, with special attention to Latin America. Second, it examines the potential use of AI-based prediction tools, discussing the institutional, data-related, and ethical challenges such implementation entails. Third, using two decades of administrative data from Argentinas prison system, it applies six machine learning models to predict reoffending. The analysis identifies economic offenses and age at incarceration as the strongest predictors, while geographic indicators also play a role, reflecting the spatial clustering of repeat offenders across prisons. The findings suggest that routinely collected prison-level information, often underutilized, can enable reasonably accurate risk prediction and guide effective rehabilitation and prison management strategies. |
| JEL: | K40 C50 I30 |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:idb:brikps:14489 |
| By: | Alexander Sheppert |
| Abstract: | Overfitting remains a critical challenge in data-driven financial modeling, where machine learning (ML) systems learn spurious patterns in historical prices and fail out of sample and in deployment. This paper introduces the GT-Score, a composite objective function that integrates performance, statistical significance, consistency, and downside risk to guide optimization toward more robust trading strategies. This approach directly addresses critical pitfalls in quantitative strategy development, specifically data snooping during optimization and the unreliability of statistical inference under non-normal return distributions. Using historical stock data for 50 S&P 500 companies spanning 2010-2024, we conduct an empirical evaluation that includes walk-forward validation with nine sequential time splits and a Monte Carlo study with 15 random seeds across three trading strategies. In walk-forward validation, GT-Score improves the generalization ratio (validation return divided by training return) by 98% relative to baseline objective functions. Paired statistical tests on Monte Carlo out-of-sample returns indicate statistically detectable differences between objective functions (p |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00080 |
| By: | Simon Deakin; Linda Shuku |
| Abstract: | The use of natural language processing (NLP) and machine learning (ML) to analyse the structure of legal texts is a fast-growing field. While much attention has been devoted to the use of these techniques to predict case outcomes, they have the potential to contribute more broadly to research into the nature of legal reasoning and its relationship to social and economic change. In this paper, we use recently developed NLP and ML methods to test the claim that judicial language is systematically shaped by economic shocks deriving from the business cycle and by long-run trends in the economy associated with technological change and industrial transition. Focusing on cases decided under the Anglo-Welsh poor law between the 1690s and 1830s, we show that the terminology used to describe the right to poor relief shifted over time according to economic conditions. We explore the implications of our results for the poor law, the theory of legal evolution, and socio-legal research methods. |
| Keywords: | Law and computation, poor law, legal evolution, natural language processing |
| JEL: | J41 K31 N33 |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:cbr:cbrwps:wp546 |
| By: | Kevin Mojica (School of Government, Universidad de los Andes) |
| Abstract: | This study analyzes data bias in an unsupervised learning algorithm designed to identify the risk of corruption in public procurement of Colombia. The employed algorithm is a two-stage clustering model used to segment electronic contracts based on variables indicating corruption risk. The objective was to develop an early warning tool for corruption in the Programa de Alimentación Escolar (PAE) procurement, utilizing data from the Sistema Electrónico para la Contratación Pública (SECOP). Although the results demonstrate the potential of artificial intelligence algorithms for detecting corruption risks, they also reveal significant limitations in their practical implementation, attributable to data availability and quality deficiencies. Specifically, biases of representation, measurement, and omitted variables were identified, affecting the algorithm's reliability. The study provides a detailed analysis of these biases, assessing their impact on the algorithm's performance, and emphasizes the importance of recognizing and addressing biases during the development of such initiatives. Finally, recommendations are presented to improve the quality of data in SECOP, aiming to enhance the reliability and accuracy of these algorithms in future developments. |
| Keywords: | data bias; unsupervised learning; Artificial Intelligence; public procurement; AI, fairness. |
| Date: | 2025–03–13 |
| URL: | https://d.repec.org/n?u=RePEc:col:000547:022180 |
| By: | Kilic, Talip; Letta, Marco; Montalbano, Pierluigi; Petruccelli, Federica |
| Abstract: | This paper proposes a new resilience index, CLARE (Causal machine Learning Approach to Resilience Estimation), which is rooted in an impact evaluation framework and causal machine learning algorithms applied to longitudinal household survey data. The indicator is model-agnostic, data-driven, scalable, and normatively anchored to wellbeing thresholds, and can be either shock-specific or a general-purpose resilience metric. The paper provides an empirical demonstration of constructing the CLARE resilience index, leveraging more than 28, 000 household observations from 19 nationally representative, longitudinal, multi-topic surveys that were implemented by the national statistical offices in Malawi, Nigeria, Tanzania, and Uganda over 2009–20 in partnership with the World Bank Living Standards Measurement Study. Although the paper centers on measuring resilience to drought, the proposed index is applicable to any type of shock. The analysis shows that CLARE outperforms existing resilience metrics and alternative approaches to predict food insecurity out-of-sample—both in the future (dynamic forecasting) and in held-out countries (cross-sectional prediction). The index can be decomposed to causally identify the relative importance of resilience capacities that can insulate populations from shocks. Thus, it can be operationalized in designing, targeting, and monitoring policies and investments that aim to strengthen resilience. CLARE’s deployment—paired with continued investments in national longitudinal survey platforms—can boost the effectiveness of early-warning systems and resilience-building interventions, while allowing the transfer of resilience policy advice from data-rich contexts to data-poor environments that may not immediately provide the requisite longitudinal survey data for index estimation. |
| Date: | 2026–01–12 |
| URL: | https://d.repec.org/n?u=RePEc:wbk:wbrwps:11292 |
| By: | Joshua S. Gans |
| Abstract: | Machine learning systems embed preferences either in training losses or through post-processing of calibrated predictions. Applying information design methods from Strack and Yang (2024), this paper provides decision-problem-agnostic conditions under which separation—training preference-free and applying preferences ex post is optimal. Unlike prior work that requires specifying downstream objectives, the welfare results here apply uniformly across decision problems. The key primitive is a diminishing-value-of-information condition: relative to a fixed (normalised) preference-free loss, preference embedding makes informativeness less valuable at the margin, inducing a mean-preserving contraction of learned posteriors. Because the value of information is convex in beliefs, preference-free training weakly dominates for any expected-utility decision problem. This provides theoretical foundations for modular AI pipelines that learn calibrated probabilities and implement asymmetric costs through downstream decision rules. However, separation requires users to implement optimal decision rules. When cognitive constraints bind—as documented in human-AI decision-making—preference embedding can dominate by automating threshold computation. These results provide design guidance: preserve optionality through postprocessing when objectives may shift; embed preferences when decision-stage frictions dominate. |
| JEL: | C45 C53 D81 D82 D83 |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:34780 |
| By: | Neetu Garg; A. S. V. Ravi Kanth |
| Abstract: | This paper implements an efficient numerical algorithm for the time-fractional Black-Scholes model governing European options. The proposed method comprises the Crank-Nicolson approach to discretize the time variable and exponential B-spline approximation for the space variable. The implemented method is unconditionally stable. We present few numerical examples to confirm the theory. Numerical simulations with comparisons exhibit the supremacy of the proposed approach. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00201 |
| By: | Kaicheng Chen; Harold D. Chiang |
| Abstract: | This paper develops an asymptotic theory for two-step debiased machine learning (DML) estimators in generalised method of moments (GMM) models with general multiway clustered dependence, without relying on cross-fitting. While cross-fitting is commonly employed, it can be statistically inefficient and computationally burdensome when first-stage learners are complex and the effective sample size is governed by the number of independent clusters. We show that valid inference can be achieved without sample splitting by combining Neyman-orthogonal moment conditions with a localisation-based empirical process approach, allowing for an arbitrary number of clustering dimensions. The resulting DML-GMM estimators are shown to be asymptotically linear and asymptotically normal under multiway clustered dependence. A central technical contribution of the paper is the derivation of novel global and local maximal inequalities for general classes of functions of sums of separately exchangeable arrays, which underpin our theoretical arguments and are of independent interest. |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.11333 |
| By: | Jan Ditzen; Erkal Ersoy; Haoyang Li; Francesco Ravazzolo |
| Abstract: | This paper studies whether a small set of dominant countries can account for most of the dynamics of regional oil demand and improve forecasting performance. We focus on dominant drivers within the OECD and a broad GVAR sample covering over 90\% of world GDP. Our approach identifies dominant drivers from a high-dimensional concentration matrix estimated row by row using two complementary variable-selection methods, LASSO and the one-covariate-at-a-time multiple testing (OCMT) procedure. Dominant countries are selected by ordering the columns of the concentration matrix by their norms and applying a criterion based on consecutive norm ratios, combined with economically motivated restrictions to rule out pseudo-dominance. The United States emerges as a global dominant driver, while France and Japan act as robust regional hubs representing European and Asian components, respectively. Including these dominant drivers as regressors for all countries yields statistically significant forecast gains over autoregressive benchmarks and country-specific LASSO models, particularly during periods of heightened global volatility. The proposed framework is flexible and can be applied to other macroeconomic and energy variables with network structure or spatial dependence. |
| Date: | 2026–02 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.01963 |
| By: | Avi Arora; Ritesh Malpani |
| Abstract: | Prediction markets offer a natural testbed for trading agents: contracts have binary payoffs, prices can be interpreted as probabilities, and realized performance depends critically on market microstructure, fees, and settlement risk. We introduce PredictionMarketBench, a SWE-bench-style benchmark for evaluating algorithmic and LLM-based trading agents on prediction markets via deterministic, event-driven replay of historical limit-order-book and trade data. PredictionMarketBench standardizes (i) episode construction from raw exchange streams (orderbooks, trades, lifecycle, settlement), (ii) an execution-realistic simulator with maker/taker semantics and fee modeling, and (iii) a tool-based agent interface that supports both classical strategies and tool-calling LLM agents with reproducible trajectories. We release four Kalshi-based episodes spanning cryptocurrency, weather, and sports. Baseline results show that naive trading agents can underperform due to transaction costs and settlement losses, while fee-aware algorithmic strategies remain competitive in volatile episodes. |
| Date: | 2026–01 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2602.00133 |