770 résultats pour « Autre »

'Egalitarian pooling and sharing of longevity risk', a.k.a. 'The many ways to skin a tontine cat'

Experts agree on the societal benefits of pooling longevity risk through annuities and pensions. While pooling reduces the upfront capital needed for a secure income, challenges arise when participants vary in wealth and health. This paper proposes a model for distributing income in diverse longevity-risk pools, emphasizing the role of social cohesion.

Partially Law‑Invariant Risk Measures

The study introduces partial law invariance, a novel concept extending law-invariant risk measures in decision-making under uncertainty. It characterizes partially law-invariant coherent risk measures with a unique formula, deviating from classical approaches. Strong partial law invariance is introduced, proposing new risk measures like partial versions of Expected Shortfall for risk assessment under model uncertainty.

Tort Law as a Tool for Mitigating Catastrophic Risk from Artificial Intelligence

This paper addresses the inadequacy of the current U.S. tort liability system in handling the catastrophic risks posed by advanced AI systems. The author proposes punitive damages to incentivize caution in AI development, even without malice or recklessness. Additional suggestions include recognizing AI as an abnormally dangerous activity and requiring liability insurance for AI systems. The paper concludes by acknowledging the limits of tort liability and exploring complementary policies for mitigating catastrophic AI risk.

Can Words Reveal Fraud? A Lexicon Approach to Detecting Fraudulent Financial Reporting

The study introduces a fraud lexicon and a Balanced Random Forest classifier for detecting fraudulent financial reporting. The classifier, utilizing the fraud lexicon as a feature set, demonstrates strong accuracy in predicting fraud across multiple samples from 2000 to 2017, outperforming random guessing by 40 to 48 percent. The fraud lexicon proves valuable for "bag-of-words" analysis, benefiting researchers, practitioners, auditors, regulators, and investors in enhancing fraud risk assessment procedures.

Robust Estimation of Pareto’s Scale Parameter from Grouped Data

The paper introduces a new robust estimation technique, the Method of Truncated Moments (MTuM), tailored for estimating the tail index of a Pareto distribution from grouped data. It addresses limitations in existing methods for grouped loss severity data, providing inferential justification through the central limit theorem and simulation studies.

Rethinking Experts’ Perception of Risk in Anti‑Money Laundering Risk Assessment

“This paper explores the factors that impact how experts perceive the risk of money laundering during Anti-Money Laundering (AML) risk assessments. To achieve this, we utilized two different exploratory methods... The study’s results suggest that experts heavily rely on their organization’s risk response and are often influenced by preconceived notions or fear.”

Understanding Polycrisis: Definitions, Applications, and Responses

"We envision a polycrisis as a state in which multiple, macroregional, ecologically embedded, and inexorably interconnected systems face high – and advancing – risk across socioeconomic, political, and other dimensions. We differentiate the term from others widely used, such as cascading disasters, compound disasters, and recurring acute disasters."

Open banking, shadow banking and regulation

Open banking creates diverse models: competitive and monopolistic banks. Policy changes impacting relative profitability lead banks to shift types. Increased capital requirements favor competitive banks, potentially raising system risk. Deposit rate ceilings can increase risk by promoting growth in the riskier competitive sector. Introducing a shadow banking sector benefits monopolistic banks, reducing overall system risk.

Deep Generative Modeling for Financial Time Series with Application in VaR: A Comparative Review

This paper explores risk factor distribution forecasting in finance, focusing on the widely used Historical Simulation (HS) model. It applies various deep generative methods for conditional time series generation and proposes new techniques. Evaluation metrics cover distribution distance, autocorrelation, and backtesting. The study reveals HS, GARCH, and CWGAN as top-performing models, with potential future research directions discussed.