Propagation of Uncertainties through Robust Simulation and Future Research

Abstract eng:
The fundamental question that has stimulated past papers is “How does one account for uncertainties in catastrophe risk analysis?” The additional question raised in this paper is “What role does propagation of uncertainty play in accounting for uncertainties?” We first illustrate how robust simulation helps to address these two questions. We use two credible but divergent accounts of earthquake risk models as illustrated through their application to industry-wide California insurance properties. We contrast the loss distribution results of these two earthquake risk models. After evaluating how these divergent approaches address the questions above, we then proceed with an illustration of how more advanced geo-science techniques may be considered in the process of building catastrophe risk models. We examine a project that uses kinematic models of the 1994 Northridge earthquake, and propagates seismic waves without uncertainties these events through the earth at depth. Based on this project, we address some of the possible pros and cons of extending this type of effort to the creation of kinematic source models that may better reflect what to expect in future earthquakes. Robust simulation requires more numerical accounting of uncertainties; this research project illustrates how one must qualify results in light of the constant challenges to our current knowledge.

• Export as: BibTeX | MARC | MARCXML | DC | EndNote | NLM | RefWorks
• Add to your basket