|
on Big Data |
| By: | Wayne Gao; Sukjin Han; Annie Liang |
| Abstract: | Large language models (LLMs) are increasingly used to predict human behavior. We propose a measure for evaluating how much knowledge a pretrained LLM brings to such a prediction: its equivalent sample size, defined as the amount of task-specific data needed to match the predictive accuracy of the LLM. We estimate this measure by comparing the prediction error of a fixed LLM in a given domain to that of flexible machine learning models trained on increasing samples of domain-specific data. We further provide a statistical inference procedure by developing a new asymptotic theory for cross-validated prediction error. Finally, we apply this method to the Panel Study of Income Dynamics. We find that LLMs encode considerable predictive information for some economic variables but much less for others, suggesting that their value as substitutes for domain-specific data differs markedly across settings. |
| Date: | 2026–01–30 |
| URL: | https://d.repec.org/n?u=RePEc:bri:uobdis:26/835 |
| By: | Saeed Varasteh Yazdi (EM - EMLyon Business School) |
| Abstract: | Reliable demand forecasting is crucial for effective supply chain management, where inaccurate forecasts can lead to frequent out-of-stock or overstock situations. While numerous statistical and machine learning methods have been explored for demand forecasting, reinforcement learning approaches, despite their significant potential, remain little known in this domain. In this paper, we propose a multi-agent deep reinforcement learning solution designed to accurately predict demand across multiple stores. We present empirical evidence that demonstrates the effectiveness of our model using a real-world dataset. The results confirm the practicality of our proposed approach and highlight its potential to improve demand forecasting in retail and potentially other forecasting scenarios. |
| Keywords: | reinforcement learning, multi-agent systems, demand forecasting |
| Date: | 2025–07–31 |
| URL: | https://d.repec.org/n?u=RePEc:hal:journl:hal-05656779 |
| By: | Tobias Lausser; Joao Eduardo Vuolo; Rudi Zagst |
| Abstract: | This paper compares different methods for forecasting the term structure of U.S. and European zero-coupon government bonds using both traditional econometric and Machine Learning (ML) approaches. We compare classical models (e.g., Dynamic Nelson-Siegel (DNS) and Principal Component Analysis (PCA)) with different Neural Network (NN) architectures, including those inspired by the classical models, on the U.S. Treasury market and bonds issued by the European Central Bank (ECB). To enhance predictive performance, macroeconomic variables are incorporated. The findings for both markets are separately analyzed and compared. To this end, we propose a robust model evaluation framework combining statistical accuracy metrics - such as RMSE, MAE, and directional accuracy - with the economic relevance of a quantitative bond trading strategy. Results show that NNs consistently outperform traditional models in both forecasting accuracy and portfolio performance. For the U.S., the most effective approach is a direct-forecasting NN that incorporates DNS factors to reduce the dimensionality of zero-rate data and an Autoencoder (AE) to extract macroeconomic features, while for Europe, the optimal model is a factor-based NN using PCA-derived zero-rate factors without the integration of macroeconomic variables. Overall, the paper demonstrates how combining traditional modeling approaches with modern ML techniques and evaluation can improve yield curve forecasts and support applications in fixed-income portfolio construction. |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2606.26815 |
| By: | Antonella Basso (Ca’ Foscari University of Venice); Marco Corazza (Ca’ Foscari University of Venice); Lorenzo Tonon (Ca’ Foscari University of Venice) |
| Abstract: | Accurate evaluation of real estate prices represents a complex task, that holds significant relevance for the majority of the actors involved in a society, as homebuyers/sellers, investors, and policymakers. Over the last decades, in parallel with technological advancements, scholars and practitioners proposed several models and methodologies, spanning from simple hedonic regression models to sophisticated machine learning algorithms, in the attempt to improve the accuracy of price forecasts. Nonetheless, the interest towards the development of more precise and performing methodologies remains high, not only for static valuation but also, more recently, for dynamic approaches. Focusing on dynamic valuation, we turn the attention towards the Recurrent Neural Network (RNN) models: while widely tested for stock prices forecast, RNNs have seen limited to no application in the domain of real estate valuation. In this study, using official data on prices of Italian residential properties, we propose and implement two different architectures for RNN models to perform both price prediction and relevance analysis with respect to the explanatory variables. Differently from other studies, we utilize data for the whole Italian territory, focusing, rather than on individual properties' valuation, on a range of prices for groups of properties identified by house type, geographical position and state of conservation. Specifically, the output variables of our analysis are represented by the minimum and maximum prices that determine this range. The results of our analysis indicate that a Standard Recurrent Neural Network forecasts well the prices in small municipalities (with less than 15 000 inhabitants), while a Double-Input Layer RNN, with an input layer dedicated to static features and a separate input layer for dynamic features, performs better in medium-sized (with a number of inhabitants between 15 000 and 50 000 inhabitants) and large municipalities (with more than 50 000 municipalities). |
| Keywords: | Real Estate valuation, House price forecast, Recurrent Neural Networks, Machine Learning, LSTM Networks |
| JEL: | C45 G17 R31 R32 |
| Date: | 2026 |
| URL: | https://d.repec.org/n?u=RePEc:ven:wpaper:2026:22 |
| By: | Torben S. D. Johansen (University of Southern Denmark); Julius Koschnick (University of Southern Denmark); Christian Vedel (University of Southern Denmark) |
| Abstract: | Machine learning (ML) has rapidly transformed economic history, lowering costs of digitization, data linkage, and imputation, and making information in historical text usable at scale. This paper offers a practical guide to using these tools well. However, ML tools have also created new problems. Prediction errors are often systematically correlated with covariates of interest, so even highly accurate models can distort and sometimes reverse coefficients, and standard validation cannot detect this. Given that ML tools often perform worse for historical data, this problem is especially severe for the field of economic history. We also identify a solution to this problem. We show that recent debiasing methods can correct such bias for a wide class of applications, using a small, randomly sampled set of expert-coded labels while retaining the efficiency of large-scale prediction. We organize the field with a taxonomy of three ML tasks, survey the literature along it, and indicate where debiasing applies and where validation against proxies remains the only recourse. We close with best-practice guidance on digitization, model choice, and reproducibility. |
| Keywords: | Machine learning, Large language models, bias correction, economic history |
| JEL: | N01 C55 C80 C81 C82 C45 |
| Date: | 2026–07 |
| URL: | https://d.repec.org/n?u=RePEc:hes:wpaper:0306 |
| By: | Bergh, Andreas (Lund University and); Anzén Ekman, Christina (Lund University); Moricz, Sara (Sensative) |
| Abstract: | How do municipalities use social media? For place marketing, civic information, or dialogue with citizens? We address this question by classifying 35, 930 tweets from 15 municipal Twitter accounts in the Skåne region of Sweden, using a machine learning algorithm trained on manually annotated data. Tweets are classified along two dimensions: whether they contain place marketing content and whether they contain civic information, yielding four categories. Our findings show that place marketing content declined substantially over the period studied, from nearly 40 percent of tweets in 2009 to just above 20 percent in 2018. In contrast, civic information and direct dialogue with citizens increased. These results suggest that the rise of social media has not trapped municipalities in a zero-sum place marketing arms race. Instead, municipalities appeared to be using Twitter increasingly as a channel for civic communication, with implications for how scholars and practitioners understand the evolving role of social media in place branding. |
| Keywords: | Place marketing; Place branding; Social media; Municipalities; Machine learning; Twitter; Sweden |
| JEL: | H83 M31 R58 |
| Date: | 2026–06–22 |
| URL: | https://d.repec.org/n?u=RePEc:hhs:iuiwop:1563 |
| By: | Pedro Salas Rojo (CUNEF University) |
| Abstract: | This paper aims to contribute to an understanding of the extent, nature and persistence of unfair inequality in the Asia Pacific region, building on a rich literature on the measurement of inequality of opportunity (IOp). As part of a project to build a global database of IOp, the paper uses microdata from 39 nationally representative household surveys to present IOp estimates for 14 countries that account for about three-quarters of the region’s population. We use consistent data protocols to ensure a high degree of cross-country comparability of IOp estimates. A distinguishing feature of the exercise is the use of machine learning methods to construct IOp estimates, which efficiently balances the risks of potential under- or over-fitting. The results show that, on average, nearly two-fifths of income or consumption inequality across the Asia-Pacific region represents inequality of opportunity attributable to inherited circumstances, though with wide variation across countries, ranging from about a quarter to over half. The cross-country variation in IOp is consistent with a Great Gatsby curve for the Asia-Pacific. A decomposition analysis assesses the relative contributions of different circumstances to IOp. |
| Keywords: | inequality of opportunity, economic mobility, Asia-Pacific, machine learning |
| JEL: | D31 D63 O15 |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:inq:inqwps:ecineq2026-696 |
| By: | Sukjin Han |
| Abstract: | The instrumental variables (IVs) method is a leading empirical strategy for causal inference. Finding IVs is a heuristic and creative process, and justifying its validity — especially exclusion restrictions — is largely rhetorical. We propose using large language models (LLMs) to search for new IVs through narratives and counterfactual reasoning, similar to how a human researcher would. The stark difference, however, is that LLMs can dramatically accelerate this process and explore an extremely large search space. We demonstrate how to construct prompts to search for potentially valid IVs. We contend that multi-step and role-playing prompting strategies are effective for simulating the endogenous decision-making processes of economic agents or social actors and for navigating language models through the realm of real-world scenarios. We apply our method to four well-known examples in economics: returns to schooling, demand and supply, and peer effects. Expert surveys reveal that some of the discovered IVs in each domain appear both novel and likely valid. |
| Date: | 2026–01–30 |
| URL: | https://d.repec.org/n?u=RePEc:bri:uobdis:26/833 |
| By: | Tom Gillespie (University of Galway); Ronan Lyons (Trinity College Dublin); Hannah Ortega-McCormack (Department of Economics) |
| Abstract: | Energy performance certificates are intended to incentivize residential energy efficiency investment by allowing markets to capitalize lower energy costs into higher property values. A large empirical literature documents price premiums for energy‐efficient dwellings, but these estimates may be biased upward if internal dwelling condition, which is correlated with energy ratings, is not adequately controlled for. We address this measurement problem using a large language model to extract a scalable, consistent measure of internal condition from the full text of 374, 000 residential listings in Ireland between 2018 and 2025. Because the approach requires only that properties be described in natural language, it is broadly transferable to other markets. Integrating this measure into a standard hedonic framework, we find that a one‐step improvement in Ireland's Building Energy Rating is associated with a price premium of approximately 2.4 percent in conventional specifications. Once internal condition is controlled for, this falls to around 1.5 percent ‐ meaning that models that omit internal quality may be overstating the energy efficiency premium by 60 percent. This attenuation holds across alternative efficiency measures, throughout the price distribution, and in matched transaction‐price regressions. These findings suggest that market‐based incentives for energy efficiency investment are more limited than conventional estimates imply but also that text-based quality measures derived from LLMs offer a practical tool for improving the credibility of hedonic analysis wherever listing data exist. |
| Keywords: | Energy efficiency; internal condition; housing markets |
| JEL: | C55 Q41 R31 |
| Date: | 2026–06 |
| URL: | https://d.repec.org/n?u=RePEc:tcd:tcduee:tep1226 |
| By: | Betz, Felix; Bofinger, Peter; Dix, Jonas; Streit, Leonie |
| Abstract: | One of the central challenges in identifying the causal effects of monetary policy is the inherent endogeneity of its conduct. This paper introduces a novel identification strategy that leverages LLMs to detect monetary policy shocks from newspaper coverage following European Central Bank (ECB) policy decisions. Based on a dataset of 7, 620 articles from eleven major European newspapers, we classify each policy decision as unexpectedly restrictive, unexpectedly expansionary, or as expected. The resulting narrative-based surprise series captures immediate post-announcement perceptions and shows a close alignment with established High Frequency Identification (HFI) measures with notable exceptions during times of financial turmoil. We subsequently analyze the potential influence of the information effect on our series and find that the majority of identified surprises are unlikely to be driven by information effects. |
| Keywords: | Monetary policy shocks; Natural language processing; Large Language Models |
| JEL: | E52 E58 C88 |
| Date: | 2026–04 |
| URL: | https://d.repec.org/n?u=RePEc:cpr:ceprdp:21390 |