Risks MDPI

19/08/2026

🙌 - 19 August

✔️ From Compliance to Resilience: Integrating Additional Risk Factors into AML Business Risk Assessments

✍️ by Yelena Popova, Olegs Cernisevs, PhD and Evita Kalmane-Pivkina

Are AML (anti-money laundering) risk assessments missing the bigger picture? This study addresses a critical gap in Business Risk Assessments (BRAs) by proposing and validating a structured approach for integrating institution‑specific "additional factors." Using PLS‑SEM on data from five European financial institutions, it finds that governance, ICT risk, operational, ESG, HR, and regulatory compliance drive 80.9% of AML vulnerability—with governance and ICT emerging as the most critical levers for resilience.

👉 https://brnw.ch/21x52Nk

Risk-based approaches constitute the cornerstone of contemporary anti-money laundering (AML) regulatory frameworks, requiring institutions to conduct Business Risk Assessments (BRAs). While the customer risk assessment component is well-defined across key regulatory dimensions, the “additional fac...

19/08/2026

🙌 - 19 August

✔️ Risk‑Sensitive Performance Evaluation of Life Insurance Markets in EU and EEA Countries: A MPSI–CoCoSo Approach

✍️ by Neylan Kaya, Aslıhan Ersoy Bozcuk, Güler Ferhan Ünal Uyar, Münevver Sena Özden, Mustafa Terzioğlu, Burçin Tutcu and Hasan Talaş

Which European life insurance markets truly lead—and what drives their success? Using MPSI and CoCoSo methods on EIOPA data, this study evaluates EU/EEA life insurance performance across eight financial stability, profitability, growth, and risk criteria. ROE emerges as the decisive factor, with Cyprus, Hungary, and Iceland showing significant positive differentiation. The findings offer a risk‑sensitive benchmark for regulators and industry stakeholders.

👉 https://brnw.ch/21x52nA

The life insurance sector plays a critical role in the financial stability of countries due to its long-term liability structure and strong interaction with the financial system. The aim of this study is to evaluate the performance of the life insurance sector in the EU and EEA countries using a mul...

18/08/2026

🙌 - 18 August

✔️ Crisis‑Regime Dynamic Volatility Spillovers in U.S. Commodity Markets: A Bayesian Mixture‑Identified SVAR Approach

✍️ by Xinyan Deng, Kentaka Aruga and Chaofeng Tang

What happens to commodity markets when crises collide? This study introduces BSVAR‑MIX, a Bayesian mixture‑based SVAR model that captures regime‑dependent spillovers across U.S. food, energy, and metals markets (2008‑2025). Gold and oil dominate systemic transmission—but even safe havens may amplify stress during extreme events. The findings reveal a complex, crisis‑shaped web of cross‑commodity risk.

👉 https://brnw.ch/21x50Nf

18/08/2026

✨ Sharing 🏆 Editor's Choice - 18 August

🎯 Title: A Novel Federated Transfer Learning Framework for Credit Card Fraud Detection Under Heterogeneous Data Conditions

✍ by Yutong Chen, Southwestern University of Finance and Economics, Kai Zhang, Zhejiang University, Hangyu Zhu, University of Illinois Urbana-Champaign and Zihao Qiu, Sichuan University

Banks face a dilemma: share data to catch fraud—or protect privacy? This Editor's Choice study offers a way out.

Introducing FED-SPFD, a federated learning framework that tackles data heterogeneity across institutions without compromising privacy. The model uses share–private segmentation to align shared features while keeping sensitive data local. A "private autoencoder + Gaussian alignment" mechanism stabilizes training.

Tested on real-world Kaggle data, FED-SPFD significantly improves recall over state-of-the-art methods—offering a compliant cross-institutional risk collaboration tool that financial institutions can actually use.

👉 Link: https://brnw.ch/21x50uG

17/08/2026

🙌 - 17 August

✔️ Board of Directors' Characteristics, Political Connection and Risk Disclosure: Evidence from an Emerging Market Context

✍️ by Ahmad Farhan Alshira'h

What makes a company more transparent about its risks—and what holds it back? This study of 900 Jordanian firm‑year observations finds board expertise boosts risk disclosure, while CEO duality undermines it. But political connections quietly reshape the entire equation. The findings reveal a complex interplay of governance and influence in emerging markets.

👉 https://brnw.ch/21x4ZbU

This research examines Jordanian risk disclosure policies and how board size, meeting frequency, CEO duality, and board expertise affect them, exploring how political ties moderate the link between board features and risk disclosure. In 2014–2023, the research examined 90 non-financial enterprises...

17/08/2026

🙌 - 17 August

✔️ Do Uncertainty and Action Shocks Affect G7 Stock Market Synchronisation? DCC‑GARCH Evidence from the 2024 U.S. Election and the Reciprocal Tariffs Announcement

✍️ by Katarzyna Czech, PhD and Michał Wielechowski

Using DCC‑GARCH on G7 stock indices (2010‑2025), this study distinguishes two shock types: the 2024 U.S. election (uncertainty shock) temporarily decoupled European markets from the U.S., while the April 2025 reciprocal tariff announcement (action shock) significantly increased correlations across all G7‑USA pairs—with North America and Europe showing the largest synchronisation. Two shocks, two very different market responses.

👉 https://brnw.ch/21x4YY6

Exogenous shocks can affect equity markets by changing volatility and cross-market co-movement. This study examines how two U.S.-centred events, treated as different shock types, influence time-varying conditional correlations between the U.S. stock market and other G7 markets. The uncertainty shock...

14/08/2026

🙌 - 14 August

✔️ Predicting Stock Market Risk Using Machine Learning Classification Models

✍️ by Seol-Hyun Noh

Can simple models outperform complex ones in predicting market crashes? Using ten years of KOSPI 200 data (2015–2024), this study labels sharp declines (returns below the 5th percentile over 100 days) and tests nine classifiers. Logistic Regression emerges as the top performer across accuracy, F1, and AUC—proving that sometimes, simpler is better for crisis detection.

👉 https://brnw.ch/21x4Vmw

This study aims to predict stock market risk and improve preparedness for potential economic crises by identifying sharp declines in stock returns using classification-based machine learning models. Using ten years of KOSPI 200 index data (2015 to 2024), a daily return series was constructed. A day....

13/08/2026

🙌 - 13 August

✔️ Geoeconomic Fragmentation and Market Decoupling: A Time–Frequency Anatomy of Oil–Ruble Volatility Spillovers (2020–2025)

✍️ by Erdost Torun, Erhan DEMİRELİ and Simon Grima

Using wavelet coherence and predictive information flow, this study dissects WTI‑Ruble volatility spillovers across time‑frequency domains. The 2020 shock saw negative spillovers with WTI leadership; the 2022 Russia‑Ukraine war triggered strong positive contagion. But by 2024, a structural decoupling emerged—the ruble no longer behaves as a traditional petro‑currency, signaling new risks for reserve managers and liquidity markets.

👉https://brnw.ch/21x4TUb

The interaction between crude oil prices and exchange rates is central to understanding global financial stability and macro-economic balances. Contrary to traditional static analyses, the heterogeneous market hypothesis argues that market participants have different time horizons and that multi-sca...

13/08/2026

🙌 ✨ - 13 August

✔️ Advanced Insurance Risk Modeling for Pseudo‑New Customers Using Balanced Ensembles and Transformer Architectures

✍️ by Finn L. Solly, Raquel Soriano González, Angel A. Juan and Antoni Guerrero Portolés

Classifying insurance customers with no prior history is uniquely challenging—class imbalance, heavy‑tailed losses, and operational constraints limit traditional methods. This study tests two profit‑driven approaches: a balanced bagging ensemble and a lightweight Transformer. Both outperform the baseline (p < 0.001), with the ensemble offering the best performance‑efficiency‑interpretability trade‑off for regulated environments, while the Transformer shows stronger robustness under data perturbations.

👉 https://brnw.ch/21x4TBv

In insurance portfolios, classifying customers without a prior history at a given company is particularly challenging due to the absence of historical behavior, extreme class imbalance, heavy-tailed loss distributions, and strict operational constraints. Traditional machine learning approaches, incl...

12/08/2026

🙌 - 12 August

✔️ Modeling Structural Deviation in 10-K Risk Factors: A Semantic Anomaly Detection and Explainable AI Approach

✍️ by Fang Sun, Shuangjiang He, Ruiqi Wang, Lingyun Ke, Hongyu Shen and Qiuyue Liao

Can AI detect when a company's risk narrative silently shifts? This study introduces a structural semantic deviation framework using sentence embeddings and isolation‑based anomaly detection on 10‑K filings. In a Wells Fargo/JPMorgan case study around the 2016 regulatory shock, embedding‑based trajectories reveal narrative reconfiguration where lexical metrics show no clear break—with SHAP highlighting semantic distance, litigation emphasis, and disclosure contraction as key deviation drivers.

👉 https://brnw.ch/21x4RQF

This study presents an exploratory methodological framework for examining structural changes in regulatory risk disclosure using sentence embeddings, multivariate anomaly detection, and explainable artificial intelligence. Prior research typically relies on dictionary-based word frequencies, tone in...