Finding the right model for the decision: How to evaluate catastrophe models for their intended use

About this webinar

Catastrophe modelling has matured considerably. There are now dozens of modelling companies spanning everything from global probabilistic catastrophe models to highly specialised hazard, vulnerability and climate products.

This diversity is a strength. Different models can offer different perspectives on risk, helping organisations better understand uncertainty, challenge assumptions and make more informed decisions.

But it also makes a familiar question harder to answer: which model is right for which decision?

There is no simple answer. Evaluation needs to be sufficiently rigorous to support confident and defensible decisions, while recognising that not every model needs to be assessed in the same way or to the same depth. The right approach will depend on the model, the intended use, the materiality of the decision and the level of uncertainty involved.

For those making these choices, the consequences are practical, commercial and technical. Frameworks designed for like-for-like comparison can create false consistency when applied to fundamentally different models. Evaluating every model separately, meanwhile, can become slow and expensive.

Governance expectations are also increasing. Firms must be able to explain not only what a model produces, but why it was selected, where its limitations sit and how those limitations should influence its use.

This webinar will examine how insurers can move beyond searching for a single “best” model and build confidence that the right model is being used for the right purpose. The panel will consider how to define criteria for model choices before an evaluation begins, compare models built on different principles and concentrate effort on the decisions where getting it wrong would have the most impact.

The discussion will focus on a practical approach to evaluation: structured enough to support consistent and defensible decisions but proportionate enough to use without turning every model choice into a major project.

Tom Philp, Chief Executive Officer and Founder of Maximum Information, and John Wardman, Chief Commercial Officer, will draw on their experience developing bespoke views of natural catastrophe risk, alongside their wider work across hazard, vulnerability and climate modelling in insurance and reinsurance.

Attendees will leave with a clearer framework for defining model appropriateness, assessing where evaluation effort is most valuable and strengthening their existing approach without over-engineering it.

In this webinar, the panel will discuss:

• Why a more diverse modelling market makes traditional like-for-like comparison less reliable, and how to avoid overlooking models that may be better suited to a particular use
• How to define what a model needs to do before evaluating it, so technical quality is considered alongside suitability for purpose
• How to compare models built on different methodologies, assumptions and objectives
• How to determine the appropriate depth of evaluation according to materiality, governance requirements and business impact
• How to strengthen an existing framework so model choices become clearer, more defensible and easier to repeat