Dynamic dependence-aware aggregation of machine-level cyber loss events

Develop accurate methods for modeling dynamic dependencies among machine-level Cyber Loss Event variables when aggregating machine-level Loss Event Frequency estimates across infrastructure levels.

Background

The LEFSA framework aggregates calibrated machine-level Cyber Loss Event probabilities to estimate infrastructure-level risk. The empirical results showed substantial aggregate overdispersion relative to the conditionally independent Bernoulli assumption, indicating residual clustering, shared latent risk factors, or other dependencies among machines.

Because machine-level dependencies may vary over time and the outcomes are discrete and rare, accurately estimating the joint distribution needed for infrastructure-level aggregation is unresolved. The paper identifies dynamic dependence modeling as a central limitation of the current empirical evaluation and motivates dependence-aware approaches, including dynamic copula-based methods, for future research.

References

While the paper proposed a generalized framework for infrastructure-level aggregation under dependence, accurate modeling of dynamic machine-level dependencies remains a challenging open problem.

— A Framework to Quantify the Probability of Future Cyber Loss Events  (2609.21717 - Peters et al., 18 Sep 2026) in Section “Limitations & operational considerations”