📈 How Actuarial Models Work (Frequency × Severity)
Actuarial models estimate future losses by analyzing how often events occur (frequency) and how costly they are when they happen (severity). This framework underpins pricing, reserving, and risk evaluation across insurance.
📘 Why This Matters
Every insurance product is built on the same core idea: expected losses. Actuaries quantify expected losses using frequency–severity models, which break risk into two components:
- Frequency — How often claims occur
- Severity — How large claims are when they occur
This structure allows actuaries to model uncertainty, forecast future outcomes, and support pricing and reserving decisions.
🔢 Frequency
Frequency measures the number of claims expected in a given period.
- Common drivers — exposure units, safety programs, industry class, geography
- Typical models — Poisson, negative binomial
- Used for — pricing, trend analysis, segmentation
💵 Severity
Severity measures the average cost of claims.
- Common drivers — inflation, litigation, medical costs, policy limits
- Typical models — lognormal, Pareto, gamma
- Used for — rate level indications, reinsurance decisions, capital modeling
📊 Expected Loss
Expected loss is the product of frequency and severity:
Expected Loss = Frequency × Severity
This simple relationship forms the foundation of actuarial pricing and reserving.
