国际学术期刊《The Review of Mathematical Economics》（https://intlpress.com/rme）于2026年8月正式出版第二期。本刊由丘成桐院士担任Honorary Editor-in-Chief，龙飞研究员担任Editor-in-Chief。

本期期刊精选以下五篇优秀论文，汇聚国内外优秀学者的前沿学术研究, 涵盖密度预测组合、机器学习在经济政策中的应用、突发公共事件与技术采纳、小型开放经济体向量误差修正模型(VECM)以及数据赋能与银行绩效等研究领域。

诚邀全球经济学界同仁关注本刊，不吝赐教，并惠赐佳作。

以下是本期论文的题目、作者列表和摘要，欢迎大家阅读分享。

1. Title: Geometric Pooling for Combining Density Forecasts

Authors: Yuyuan Wang, Jie Hu, and Yuhong Yang

DOI: 10.4310/RME.260730232358

Abstract: Density forecasting, particularly for time series data, provides a comprehensive characterization of future uncertainty by estimating the entire predictive distribution, whereas point forecasting summarizes the future by a single value. It plays an important role in many applications, including financial risk management, where predictive distributions are used to evaluate quantities such as Value-at-Risk (VaR). Although many methods are available for density forecasting, their empirical performance can differ substantially across applications, and identifying the best method in a given setting is often difficult. Forecast combination provides a practical and effective alternative, and in some cases may outperform all individual candidate forecasts. In this paper, we propose a relaxed-constraint geometric pooling method and its penalized version for combining density forecasts, and establish theoretical properties for the resulting optimized  weights. The numerical results suggest that the proposed methods work well in settings where the simplex constraint is restrictive, while also illustrating that performance gains depend on the forecasting environment and the choice of the penalty rate.

2. Title: Using Machine Learning for Prediction and Policy Analysis in Economics

Authors: Jiajing Sun,Michael Cole, and Wolfgang Karl Härdle

DOI:10.4310/RME.260730231155

Abstract: Machine learning (ML) is now part of the normal empirical toolkit of economics, but its contribution is uneven across tasks and must be judged through economic criteria. In forecasting and measurement, flexible algorithms help economists use high-dimensional administrative data, text, images, digital traces, and labor-market sequences to construct variables that were previously unobserved or observed only with delay. In causal work, ML is most useful when it estimates nuisance functions inside orthogonal, doubly robust, or sample-split procedures that protect low-dimensional treatment-effect parameters from first-stage over-fitting. In structural, financial, and market-design applications, ML expands the feasible computational frontier, but identification, equilibrium reasoning, and counterfactual discipline are still derived from economic structure. In operational settings, such as credit scoring, inspections, humanitarian targeting, central-bank communication, and social protection, predictive accuracy is only one input in a broader welfare, fairness, transparency, and governance problem. This review, therefore, reads the recent literature as a set of task-specific research designs, rather than as a generic contest between algorithms and econometrics. We pay particular attention to what representative papers actually do—their data, target, identifying or validation strategy, and substantive contribution—and to the conditions under which ML improves economic evidence, rather than merely adding complexity.

3. Title: Public Emergencies and Advanced Technology Adoption: Theory and    Firm-Level Evidence

Authors: Shihan Li,Qingfu Liu,Yiuman Tse,Christina Dan Wang, and Xiao Wei

DOI:10.4310/RME.260731234637

Abstract: Public emergencies reduce social welfare but may paradoxically stimulate corporate innovation through crisis-driven technological adoption. This study develops a theoretical framework that yields testable implications: exogenous shocks strengthen technology adoption incentives, especially for firms with lower initial technology endowments, while adoption responses vary with firm size. Empirically, we construct a firm-level digital transformation index through textual analysis of a multi-source media database in China, demonstrating that digital transformation enhances firm resilience by boosting capital market performance during public emergencies--particularly for medium-sized enterprises, given the costs and necessity of digital upgrading. Overall, this research contributes to the evidence that public emergencies can accelerate advanced technology adoption.

4. Title: Estimating a VECM for a Small Open Economy

Authors: Sebastian Ankargren and Johan Lyhagen

DOI:10.4310/RME.260731000003

Abstract: In economic theory, the term small open economy refers to an economy that is too small to influence the surrounding world. The surrounding world can, for this reason, be seen as exogenous relative to the economy of this small open economy. The main contribution of this paper is the proposal of how to estimate a vector error correction model with exogeneity restrictions on the long-run and short-run adjustment parameters as well as on the short-run dynamic parameters between small open economies and the surrounding world. A Monte Carlo simulation study of impulse responses shows that the proposed method is considerably more efficient compared to models that fully or partially ignore the restrictions implied by the small open economy hypothesis. Using two Swedish macroeconomic datasets, we find that there are, for some variables, large differences in impulse responses between our proposed method incorporating the restrictions and models using no or partial restrictions. As the small open economy hypothesis is in many situations uncontroversial, our method enables the incorporation of indisputable economic theory into the econometric estimation of the model.

5. Title: Data Empowerment and Bank Performance

Authors: Zhen Li,Ke Song, Shiying Wen and Jiawen Yang

DOI:10.4310/RME.260731230144

Abstract: This study examines how data empowerment affects bank performance. We use large language models to construct a domain-specific dictionary of bank data empowerment and apply it to bank annual reports and patent filings, yielding indicators that capture both the strategic orientation of data empowerment and the innovation activities that implement it. We find that data empowerment significantly enhances bank performance, and we identify three channels that account for part of this effect: growth in demand deposits, expansion of personal loan portfolios, and greater revenue diversification. The effect is stronger for stateowned banks, for banks with more concentrated ownership, and for banks facing higher economic policy uncertainty. These findings establish data as a core strategic resource in modern banking and point to the value of integrating internal and external data sources.

《The Review of Mathematical Economics》是一本跨学科国际学术期刊，目标是搭建数学应用与经济研究之间的桥梁。本刊专注发表兼具理论分析与实证研究的高水平学术论文，服务于经济学者、经济工作者与政策制定者，刊载内容涵盖经济学所有研究领域。本刊恪守学术独立，坚守学术诚信，以客观、公正、专业的评审准则对待每篇来稿，致力于打造思想交流、观点碰撞、智慧激荡的开放、理性、创新的国际一流学术交流平台。