Robust Actuarial Risk Analysis

Blanchet, José; Lam, Henry; Tang, Qihe; Yuan, Zhongyi

doi:10.1080/10920277.2018.1504686

Cited by 11 publications

(7 citation statements)

References 60 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For simplicity, the relationships among risks are modeled using a linear correlation, through a Gaussian copula hypothesis. In this way, it is possible to perform an uncertainty analysis, as defined by Blanchet et al, 41 defining a baseline model, that is, the definition of a hedging product neglecting the relationships among cyber risks and the positive asymmetry of distributions, and a metric to estimate the alternative model, that is, the vine copula, starting from the information related to the relationship between cyber risks defined by Biener et al 40 …”

Section: Numerical Applicationmentioning

confidence: 99%

Vine copula modeling dependence among cyber risks: A dangerous regulatory paradox

Carannante

D’Amato

Fersini

et al. 2023

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

Dependence among different cyber risk classes is a fundamentally underexplored topic in the literature. However, disregarding the dependence structure in cyber risk management leads to inconsistent estimates of potential unintended losses. To bridge this gap, this article adopts a regulatory perspective to develop vine copulas to capture dependence. In quantifying the solvency capital requirement gradient for cyber risk measurement according to Solvency II, a dangerous paradox emerges: an insurance company does not tend to provide cyber risk hedging products as they are excessively expensive and would require huge premiums that it would not be possible to find policyholders.

show abstract

Section: Numerical Applicationmentioning

confidence: 99%

Vine copula modeling dependence among cyber risks: A dangerous regulatory paradox

Carannante

D’Amato

Fersini

et al. 2023

Appl Stoch Models Bus & Ind

View full text Add to dashboard Cite

show abstract

“…This is a form of non-parametric sensitivity analysis and is reminiscent in spirit of the stress tests used in finance and actuarial science, e.g. [11] to protect against sudden changes and extreme uncertainty under various scenarios. In our Bayesian network context, individual model misspecification η l , l ∈ V for the model sensitivity indices I ± (f (X k ), P ; Q η l ) can take arbitrary fixed values that correspond to model perturbations associated with local sensitivity analysis (small η l ) or global sensitivity analysis (larger η l ).…”

Section: Stress Tests Ranking and Correctabilitymentioning

confidence: 99%

“…Related work is also encountered in macroeconomics, we refer to the book Hansen and Sargent [41]. Stress testing via a DRO perspective has been developed in the context of insurance risk analysis in [11]. Finally, [57] and [37] develop robust uncertainty quantification methods using different combinations of concentration inequalities and/or information divergences.…”

mentioning

confidence: 99%

Model Uncertainty and Correctability for Directed Graphical Models

Birmpa¹,

Feng²,

Katsoulakis³

et al. 2021

Preprint

View full text Add to dashboard Cite

Probabilistic graphical models are a fundamental tool in probabilistic modeling, machine learning and artificial intelligence. They allow us to integrate in a natural way expert knowledge, physical modeling, heterogeneous and correlated data and quantities of interest. For exactly this reason, multiple sources of model uncertainty are inherent within the modular structure of the graphical model. In this paper we develop information-theoretic, robust uncertainty quantification methods and non-parametric stress tests for directed graphical models to assess the effect and the propagation through the graph of multi-sourced model uncertainties to quantities of interest. These methods allow us to rank the different sources of uncertainty and correct the graphical model by targeting its most impactful components with respect to the quantities of interest. Thus, from a machine learning perspective, we provide a mathematically rigorous approach to correctability that guarantees a systematic selection for improvement of components of a graphical model while controlling potential new errors created in the process in other parts of the model. We demonstrate our methods in two physico-chemical examples, namely quantum scale-informed chemical kinetics and materials screening to improve the efficiency of fuel cells.

show abstract

“…(2017, 2019), Ghossoub (2019a, 2019b), Blanchet et al . (2019), Birghila and Pflug (2019), and Birghila et al . (2020), among others.…”

Section: Introductionmentioning

confidence: 98%

Worst-case moments under partial ambiguity

Tang

Yang²

2023

ASTIN Bull.

Self Cite

View full text Add to dashboard Cite

The model uncertainty issue is pervasive in virtually all applied fields but especially critical in insurance and finance. To hedge against the uncertainty of the underlying probability distribution, which we refer to as ambiguity, the worst case is often considered in quantifying the underlying risk. However, this worst-case treatment often yields results that are overly conservative. We argue that, in most practical situations, a generic risk is realized from multiple scenarios and the risk in some ordinary scenarios may be subject to negligible ambiguity so that it is safe to trust the reference distributions. Hence, we only need to consider the worst case in the other scenarios where ambiguity is significant. We implement this idea in the study of the worst-case moments of a risk in the hope to alleviate the over-conservativeness issue. Note that the ambiguity in our consideration exists in both the scenario indicator and the risk in the corresponding scenario, leading to a two-fold ambiguity issue. We employ the Wasserstein distance to construct an ambiguity ball. Then, we disentangle the ambiguity along the scenario indicator and the risk in the corresponding scenario, so that we convert the two-fold optimization problem into two one-fold problems. Our main result is a closed-form worst-case moment estimate. Our numerical studies illustrate that the consideration of partial ambiguity indeed greatly alleviates the over-conservativeness issue.

show abstract

Robust Actuarial Risk Analysis

Cited by 11 publications

References 60 publications

Vine copula modeling dependence among cyber risks: A dangerous regulatory paradox

Vine copula modeling dependence among cyber risks: A dangerous regulatory paradox

Model Uncertainty and Correctability for Directed Graphical Models

Worst-case moments under partial ambiguity

Contact Info

Product

Resources

About