754 résultats pour « Autre »

Properties of the entropic risk measure EVaR in relation to selected distributions

"Entropic Value-at-Risk (EVaR) ... was previously calculated explicitly only for the normal distribution. We succeeded ... to calculate EVaR for Poisson, compound Poisson, Gamma, Laplace, exponential, chi-squared, inverse Gaussian distribution and normal inverse Gaussian distribution with the help of Lambert function that is a special function, generally speaking, with two branches.”

The Ransomware Epidemic: Recent Cybersecurity Incidents Demystified

Date : Tags : , , ,
"... This review conducts a comprehensive literature review delving into recent ransomware attacks to analyze key aspects, including the targeted organizations, attack vectors, threat actors, propagation mechanisms, and the resulting business impact… this study provides valuable insights emphasizing the importance of proactive defenses to mitigate the risks posed by this growing threat."

Valuing insurance against small probability risks: A meta‑analysis

Date : Tags : , , ,
A meta-analysis of contingent valuation studies on voluntary insurance for low-probability, high-impact risks finds demand lower than expected. Stated willingness to pay (WTP) averages 87% of expected losses. Factors like loss probability information positively affect WTP, while income and age negatively influence it. Cultural and methodological factors also impact WTP.

Privacy‑Enhancing Collaborative Information Sharing through Federated Learning -- A Case of the Insurance Industry

The report highlights Federated Learning's (FL) benefits in claims loss modeling by enabling collaboration across multiple insurance datasets without data sharing. FL addresses data privacy concerns, rarity of claim events, and lack of informative factors. It enhances forecasting effectiveness while preserving data privacy, applicable beyond insurance to fraud detection and catastrophe modeling, fostering future collaborations.

Estimation of Spectral Risk Measure for Left Truncated and Right Censored Data

Monte Carlo studies are conducted to compare the proposed spectral risk measure estimator with the existing parametric and non parametric estimators for left truncated and right censored data. Based on our simulation study we estimate the exponential spectral risk measure for three data sets viz; Norwegian fire claims data set, Spain automobile insurance claims and French marine losses.

The Value‑Relevance of ESG Scores:A Fair Value Hierarchy Perspective with Evidence from European Banks

Date : Tags : , , , , ,
“… stock market investors either do not treat the ESG score as a reliable measure of ESG performance or, embracing the “overinvestment view” rather than the “risk mitigation view” of Corporate Social Responsibility, do not associate positive ESG performance to greater corporate transparency and trustworthiness.”

On Modeling Contagion in the Formation of Operational Risk Loss

“We lay a theoretical foundation for the choice of an exponential–Pareto combined distribution to model the severity of the operational risk. We derive, on a theoretical basis, the functional form of the operational risk severity distribution. The resulting loss severity distribution, in theory, is consistent with the parametric distribution that previous empirical works suggest is the best fit for loss data.”

The Challenge of Climate Risk Modelling in Financial Institutions - Overview, Critique and Guidance

Date : Tags : , , , ,
Climate change presents substantial risks to global finance, yet current methodologies for quantifying these risks are often incomplete or misleading due to complexity. Challenges include data quality, model uncertainty, and integration into risk management frameworks. Improved models are needed to accurately assess climate risks and inform stakeholders for coherent decision-making.

On the Potential of Network‑Based Features for Fraud Detection

Online transaction fraud poses significant challenges to businesses and consumers, with rule-based systems struggling to keep up. Machine learning, particularly personalized PageRank (PPR), offers promise by analyzing account relationships. Results show PPR enhances fraud detection models, providing valuable insights and stable features across datasets, improving predictive power.