|
on Computational Economics |
| By: | Howard Su; Huan-Hsin Tseng |
| Abstract: | We propose a quantum machine learning framework for approximating solutions to high-dimensional parabolic partial differential equations (PDEs) that can be reformulated as backward stochastic differential equations (BSDEs). In contrast to popular quantum-classical network hybrid approaches, this study employs the pure Variational Quantum Circuit (VQC) as the core solver without trainable classical neural networks. The quantum BSDE solver performs pathwise approximation via temporal discretization and Monte Carlo simulation, framed as model-based reinforcement learning. We benchmark VQCbased and classical deep neural network (DNN) solvers on two canonical PDEs as representatives: the Black-Scholes and nonlinear Hamilton-Jacobi-Bellman (HJB) equations. The VQC achieves lower variance and improved accuracy in most cases, particularly in highly nonlinear regimes and for out-of-themoney options, demonstrating greater robustness than DNNs. These results, obtained via quantum circuit simulation, highlight the potential of VQCs as scalable and stable solvers for highdimensional stochastic control problems. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.14612 |
| By: | Leland D. Crane; Akhil Karra; Paul E. Soto |
| Abstract: | We evaluate the ability of large language models (LLMs) to estimate historical macroeconomic variables and data release dates. We find that LLMs have precise knowledge of some recent statistics, but performance degrades as we go farther back in history. We highlight two particularly important kinds of recall errors: mixing together first print data with subsequent revisions (i.e., smoothing across vintages) and mixing data for past and future reference periods (i.e., smoothing within vintages). We also find that LLMs can often recall individual data release dates accurately, but aggregating across series shows that on any given day the LLM is likely to believe it has data in hand which has not been released. Our results indicate that while LLMs have impressively accurate recall, their errors point to some limitations when used for historical analysis or to mimic real time forecasters. |
| Keywords: | Artificial intelligence; Forecasting; Large language models; Real-time data |
| JEL: | C53 C80 E37 |
| Date: | 2025–06–25 |
| URL: | https://d.repec.org/n?u=RePEc:fip:fedgfe:2025-44 |
| By: | Charles Shaw |
| Abstract: | The double/debiased machine learning (DML) framework has become a cornerstone of modern causal inference, allowing researchers to utilise flexible machine learning models for the estimation of nuisance functions without introducing first-order bias into the final parameter estimate. However, the choice of machine learning model for the nuisance functions is often treated as a minor implementation detail. In this paper, we argue that this choice can have a profound impact on the substantive conclusions of the analysis. We demonstrate this by presenting and comparing two distinct Distributional Instrumental Variable Local Average Treatment Effect (D-IV-LATE) estimators. The first estimator leverages standard machine learning models like Random Forests for nuisance function estimation, while the second is a novel estimator employing Kolmogorov-Arnold Networks (KANs). We establish the asymptotic properties of these estimators and evaluate their performance through Monte Carlo simulations. An empirical application analysing the distributional effects of 401(k) participation on net financial assets reveals that the choice of machine learning model for nuisance functions can significantly alter substantive conclusions, with the KAN-based estimator suggesting more complex treatment effect heterogeneity. These findings underscore a critical "caveat emptor". The selection of nuisance function estimators is not a mere implementation detail. Instead, it is a pivotal choice that can profoundly impact research outcomes in causal inference. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.12765 |
| By: | Yonggeun Jung |
| Abstract: | This paper proposes a scalable framework to estimate monthly GDP using machine learning methods. We apply Multi-Layer Perceptron (MLP), Long Short-Term Memory networks (LSTM), Extreme Gradient Boosting (XGBoost), and Elastic Net regression to map monthly indicators to quarterly GDP growth, and reconcile the outputs with actual aggregates. Using data from China, Germany, the UK, and the US, our method delivers robust performance across varied data environments. Benchmark comparisons with prior US studies and UK official statistics validate its accuracy. The approach offers a flexible and data-driven tool for high-frequency macroeconomic monitoring and policy analysis. |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2506.14078 |
| By: | Dominic Zaun Eu Jones |
| Abstract: | I develop Ornithologist, a weakly-supervised textual classification system and measure the hawkishness and dovishness of central bank text. Ornithologist uses ``taxonomy-guided reasoning'', guiding a large language model with human-authored decision trees. This increases the transparency and explainability of the system and makes it accessible to non-experts. It also reduces hallucination risk. Since it requires less supervision than traditional classification systems, it can more easily be applied to other problems or sources of text (e.g. news) without much modification. Ornithologist measurements of hawkishness and dovishness of RBA communication carry information about the future of the cash rate path and of market expectations. |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.09083 |
| By: | Marco Zanotti |
| Abstract: | Given the continuous increase in dataset sizes and the complexity of forecasting models, the tradeoff between forecast accuracy and computational cost is emerging as an extremely relevant topic, especially in the context of ensemble learning for time series forecasting. To asses it, we evaluated ten base models and eight ensemble configurations across two large-scale retail datasets (M5 and VN1), considering both point and probabilistic accuracy under varying retraining frequencies. We showed that ensembles consistently improve forecasting performance, particularly in probabilistic settings. However, these gains come at a substantial computational cost, especially for larger, accuracy-driven ensembles. We found that reducing retraining frequency significantly lowers costs, with minimal impact on accuracy, particularly for point forecasts. Moreover, efficiency-driven ensembles offer a strong balance, achieving competitive accuracy with considerably lower costs compared to accuracy-optimized combinations. Most importantly, small ensembles of two or three models are often sufficient to achieve near-optimal results. These findings provide practical guidelines for deploying scalable and cost-efficient forecasting systems, supporting the broader goals of sustainable AI in forecasting. Overall, this work shows that careful ensemble design and retraining strategy selection can yield accurate, robust, and cost-effective forecasts suitable for real-world applications. |
| Keywords: | Time series, Demand forecasting, Forecasting competitions, Cross-learning, Global models, Forecast combinations, Ensemble learning, Machine learning, Deep learning, Conformal predictions, Green AI. |
| JEL: | C53 C52 C55 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:mib:wpaper:554 |
| By: | Raffaella Barone |
| Abstract: | This paper examines the relationship between non-residential property prices and various social, economic, and environmental indicators within the provinces where these properties are located. We focus on indicators from the Eni Enrico Mattei Foundation and SDSN Italia that track the 17 sustainable development goals, as well as additional factors like crime rates, per capita GDP, and sales frequency. Using a machine learning algorithm, we predicted property sale prices and applied SHapley Additive exPlanations to assess the importance of each variable. Our findings highlight the strong influence of categorical variables and SDG indicators on prices. Finally, we used causal inference to explore how policy interventions might affect property prices. |
| Keywords: | Machine Learning, Real estate market, Financial Stability, Sustainability, Crimes |
| JEL: | B4 C1 G01 R33 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:baf:cbafwp:cbafwp25238 |
| By: | Nicolas Houli\'e |
| Abstract: | I show that house prices can be modeled using machine learning (kNN and tree-bagging) and a small dataset composed of macro-economic factors (MEF), including an inflation metric (CPI), US treasury rates (10-yr), Gross Domestic Product (GDP), and portfolio size of central banks (ECB, FED). This set of parameters covers all the parties involved in a transaction (buyer, seller, and financing facility) while ignoring the intrinsic properties of each asset and encompassing local (inflation) and liquidity issues that may impede each transaction composing a market. The model here takes the point of view of a real estate trader who is interested in both the financing and the price of the transaction. Machine Learning allows for the discrimination of two periods within the dataset. Unconventional policies of central banks may have allowed some institutional investors to arbitrage between real estate returns and other bond markets (sovereign and corporate). Finally, to assess the models' relative performances, I performed various sensitivity tests, which tend to constrain the possibilities of each approach for each need. I also show that some models can predict the evolution of prices over the next 4 quarters with uncertainties that outperform existing index uncertainties. |
| Date: | 2025–04 |
| URL: | https://d.repec.org/n?u=RePEc:arx:papers:2505.09620 |
| By: | Liliana P. Calderón-Bernal (Max Planck Institute for Demographic Research, Rostock, Germany); Diego Alburez-Gutierrez (Max Planck Institute for Demographic Research, Rostock, Germany); Martin Kolk (Max Planck Institute for Demographic Research, Rostock, Germany); Emilio Zagheni (Max Planck Institute for Demographic Research, Rostock, Germany) |
| Abstract: | Estimating kinship networks is a data-intensive undertaking, typically conducted using empirical sources or demographic models. While empirical data, like population registers, provide a realistic picture, they are limited by scarcity, truncation, and survivorship bias. Demographic models, including microsimulation, require less detailed data but often minimally address population heterogeneity, family similarity, and multipartner fertility. This study assesses the validity of kinship networks derived from SOCSIM microsimulation by comparing kin counts (from grandparents to grandchildren) for Swedish cohorts born between 1915 and 2017 with register-based counts. The results show that microsimulation closely approximates mean kin numbers and reasonably reflects parity distributions. While it underestimates kin for recent cohorts unaffected by truncation, it more accurately captures kin for older cohorts missing parent–child links. These findings validate the use of microsimulation as a valuable tool for reconstructing kinship when only aggregate data are available, supporting its application in historical and projected kinship analyses. |
| Keywords: | Sweden, kinship, microsimulation, population registers |
| JEL: | J1 Z0 |
| Date: | 2025 |
| URL: | https://d.repec.org/n?u=RePEc:dem:wpaper:wp-2025-020 |
| By: | Dhanashree Somani (Department of Statistics, University of Florida, 230 Newell Drive, Gainesville, FL, 32601, USA); Rangan Gupta (Department of Economics, University of Pretoria, Private Bag X20, Hatfield 0028, South Africa); Sayar Karmakar (Department of Statistics, University of Florida, 230 Newell Drive, Gainesville, FL, 32601, USA); Vasilios Plakandaras (Department of Economics, Democritus University of Thrace, Komotini, Greece) |
| Abstract: | The objective of this paper is to forecast volatilities of the stock returns of China, France, Germany, Italy, Spain, the United Kingdom (UK), and the United States (US) over the daily period of January 2010 to February 2025 by utilizing the information content of newspapers articles-based indexes of supply bottlenecks. We measure volatility by employing the interquantile range, estimated via an asymmetric slope autoregressive quantile regression fitted on stock returns to derive the conditional quantiles. In the process, we are also able to obtain estimates of skewness, kurtosis, lower- and upper-tail risks, and incorporate them into our linear predictive model, alongside leverage. Based on the shrinkage estimation using the Lasso estimator to control for overparameterization, we find that the model with moments outperform the benchmark model that includes both own- and cross-country volatilities, but the performance of the former, is improved further when we incorporate the role of the metrics of supply constraints for all the 7 countries simultaneously. These findings carry significant implications for investors. |
| Keywords: | Supply Bottlenecks, Stock Market Volatility, Asymmetric Autoregressive Quantile Regression, Lasso Estimator, Forecasting |
| JEL: | C22 C53 E23 G15 |
| Date: | 2025–06 |
| URL: | https://d.repec.org/n?u=RePEc:pre:wpaper:202521 |
| By: | Hanming Fang; Ming Li; Guangli Lu |
| Abstract: | We decode China’s industrial policies from 2000 to 2022 by employing large language models (LLMs) to extract and analyze rich information from a comprehensive dataset of 3 million documents issued by central, provincial, and municipal governments. Through careful prompt engineering, multistage extraction and refinement, and rigorous verification, we use LLMs to classify the industrial policy documents and extract structured information on policy objectives, targeted industries, policy tones (supportive or regulatory/suppressive), policy tools, implementation mechanisms, and intergovernmental relationships, etc. Combining these newly constructed industrial policy data with micro-level firm data, we document four sets of facts about China's industrial policy that explore the following questions: What are the economic and political foundations of the targeted industries? What policy tools are deployed? How do policy tools vary across different levels of government and regions, as well as over the phases of an industry's development? What are the impacts of these policies on firm behavior, including entry, production, and productivity growth? We also explore the political economy of industrial policy, focusing on top-down transmission mechanisms, policy persistence, and policy diffusion across regions. Finally, we document spatial inefficiencies and industry-wide overcapacity as potential downsides of industrial policies. |
| JEL: | C55 L52 O25 |
| Date: | 2025–05 |
| URL: | https://d.repec.org/n?u=RePEc:nbr:nberwo:33814 |