Analytics Toolkit

The Foundations chapter established that reinsurance is a capital allocation problem driven by tail risk and structured through contracts.

• It is a capital allocation problem because every reinsurer has finite capital and must decide how to deploy it across contracts, perils, and geographies to maximise return
• It is driven by tail risk because the losses that matter most (the ones that threaten solvency) are rare, extreme events in the tail of the loss distribution
• It is structured through contracts because the contract is the atomic unit of risk transfer: every business decision, every metric, every allocation ultimately flows through contract terms applied to trial losses

This chapter gives you the analytical toolkit to work with that problem: the data structures, the distributions, and the metrics that every downstream application depends on. It is the measurement layer: how to read a loss distribution and extract the numbers that drive decisions. The next chapter, Financial Modelling, is the transformation layer: how contracts reshape those distributions.

By the end, you will be able to:

• Explain what a trial is and why trial-based simulation is the foundation of catastrophe risk analytics
• Read and construct exceedance probability (EP) curves
• Compute the core risk metrics: expected loss (EL), Value at Risk (VaR), and Tail Value at Risk (TVaR)
• Interpret what each metric says about the tail of a loss distribution
• Say why this site holds capital against TVaR rather than VaR, and work the counterexample that shows VaR failing
• Compute the metrics a contract needs beyond the core three: attachment and exhaustion probability, standard deviation, window TVaR

The chapter ships seven interactive elements: a Trial 9 timeline, the aggregate exceedance probability (AEP) curve and the EP curve explorer, a chart for each core metric, and a metrics explorer over the whole Caldera Re dataset. Four of them take inputs you can vary.

This chapter assumes college-level probability and statistics (expect quantiles, empirical distributions, a little set notation, and the floor function), but no actuarial background and no measure theory. Every formula comes with a plain-English restatement and the line of code that computes it.

We continue with the Caldera Re example. The company and its six contracts come from Foundations and the Caldera Re appendix; the trial loss data behind every number in this chapter is introduced in the trial worldview. The conventions this chapter introduces are indexed on Scope, conventions & assumptions.

Chapter outline

1. The trial worldview: What trials and events are, and why they are the foundation
2. Exceedance probability curves: What EP curves represent, how to build them, and the occurrence and aggregate forms
3. Metrics: EL, VaR, TVaR and the metrics that drive decisions

Each section builds on the previous, toward this chapter’s closing thesis: once you can measure a loss distribution, the next question is where those distributions come from and how contracts reshape them. The Financial Modelling chapter answers it, and argues that a handful of operations on trial losses can express the loss side of the core property-cat structures (quota share, catastrophe excess of loss, aggregate excess of loss) and of the programs and portfolios built from them.