A protection seller may be least able to pay when the protection it sold is most valuable. In that situation, multiplying an average exposure profile by an average default probability understates the claim at risk. This lesson identifies the dependence hidden by the independence CVA formula, measures it in a finite-state example, and explains what a dynamic model must preserve.
Learning objectives. By the end of this note you should be able to:
- express wrong-way risk through default-conditioned exposure;
- calculate the error in the product-of-means approximation;
- distinguish counterparty dependence from reference credit risk;
- design a controlled dependence stress without changing marginal risks.
In this lesson
- The conditioning event changes the exposure
- Derive the independence error
- A two-regime credit case
- What must move together in a dynamic model
- Collateral does not erase the dependence question
- A validation experiment that answers the right question
Before you start
The full lesson, with worked examples and exercise solutions, is for members.
Read the full lessonNew to SelfMadeQuant.com? See membership options or browse the Counterparty Credit Risk and CVA syllabus.