Nos derniers articles

This research evaluates different regression models to predict #flood-induced #insuranceclaims, using the #us #national #floodinsurance Program (#nfip) dataset from 2000 to 2020. The models studied include #neuralnetworks (Conditional Generative Adversarial Networks), #decisiontrees (Extreme Gradient Boosting), and #kernel-based regressors (#gaussian Process). The study identifies key predictors for regression, highlighting factors that influence flood-related financial damages.

The #creditsuisse #coco wipeout occurred when the #finma announced that the contingent convertible bonds that were part of the Credit Suisse Additional #tier1 (AT1) #regulatory capital had been written off.FINMA’s decision creates a healthy precedent: restoring #financialdiscipline in AT1 #bondmarkets by reminding investors that their investment is exposed to #creditrisk and that #duediligence is advised before investing in these products.

"We use #naturallanguageprocessing to #measure #supplychainrisk (#scr) faced by #us firms, as expressed in narratives of quarterly earnings conference calls."

"#ifrs17 introduces the concept of a #riskadjustment that compensates #insurers for the #uncertainty about the amount and timing of the cash flows that arise from #nonfinancial#risks. The method for its calculation is not prescribed and several options are emerging, including #var and cost of #capital."

"This paper presents an intellectual exchange with #chatgpt, … , about correlation pitfalls in #riskmanagement. … Our findings indicate that ChatGPT possesses solid knowledge of basic and mostly non-technical aspects of the topic, but falls short in terms of the mathematical goring needed to avoid certain pitfalls or completely comprehend the underlying concepts."

"We show that classical #insurance #models based on some compound distributions can well predict #information #leakage by #cyberincidents with reducing the computational cost thanks to the model’s simplicity."

This paper addresses the challenges associated with the adoption of #machinelearning (#ml) in #financialinstitutions. While ML models offer high predictive accuracy, their lack of explainability, robustness, and fairness raises concerns about their trustworthiness. Furthermore, proposed #regulations require high-risk #ai systems to meet specific #requirements. To address these gaps, the paper introduces the Key AI Risk Indicators (KAIRI) framework, tailored to the #financialservices industry. The framework maps #regulatoryrequirements from the #euaiact to four measurable principles (Sustainability, Accuracy, Fairness, Explainability). For each principle, a set of statistical metrics is proposed to #measure, #manage, and #mitigate #airisks in #finance.

"This paper introduces the multivariate range Value-at-Risk (MRVaR) and multivariate range covariance (MRCov) as #risk#measures for #riskmanagement in #regulation and investment… Frequently-used cases in industry, such as normal, student-t, logistic, Laplace, and Pearson type VII distributions, are presented with numerical examples."

#crisis #riskmanagement"The existing data show that #political #crises make #economiccrises crises more likely, so that, as suggested by the concept of #polycrisis, feedback between non-economic crises and economic crises can be important, but there is no comparable evidence for #climate events."

"... we propose applying the #risk categories to specific #ai #scenarios, rather than solely to fields of application, using a #riskassessment #model that integrates the #aia [#eu #aiact] with the risk approach arising from the Intergovernmental Panel on Climate Change (#ipcc) and related literature. This model enables the estimation of the magnitude of AI risk by considering the interaction between (a) risk determinants, (b) individual drivers of determinants, and (c) multiple risk types. We use large language models (#llms) as an example."