Week 5

In this lecture we start considering some important proprietary models we can use to estimate the probability of default (PD) of a company. All these models belong to the IRB class.

If you do not remember what an internal-rating based model is, then go back and check what we have seen in lecture 1, 2 and 4.

We will start with one of the most important models: Merton’s model.
Merton’s model is not really a proprietary model, but rather an academic one, strictly linked to the famous Black-Scholes formula. However, it is the starting point of many proprietary models, hence we need it.

Introduced in the 1970s, this model is named after Prof. Dr. Robert Merton, an eminent financial scholar. Merton’s model represents the prototype of structural models of default.

Today, given its somehow restrictive assumptions, Merton’s model is not the most used model in real-life business. Nevertheless it is a fundamental benchmark for both scholars and practitioners. Many industry models derive from Merton’s original construction.

One of these models is Moody’s KMV model, property of Moody’s Analytics. This will be the other model we will consider together during this week.

The KMV model tries to overcome some of the points of weakness of Merton’s model, by making use of some more realistic assumptions, mostly based on the empirical observation of real-life data.

This lecture contains: