Simulating the Credit Loss Distribution

SRICHANDER RAMASWAMY, PhD

Senior Economist, Bank for International Settlements, Basel, Switzerland

Abstract: Monte Carlo methods have become a valuable computational tool in modern finance as the increased availability of powerful computers has enhanced their efficiency. A particularly useful feature of Monte Carlo methods is that their computational complexity increases linearly with the number of variables. Moreover, they are flexible and easy to implement for a range of distributional assumptions for the underlying variables that influence the outcomes of interest. Monte Carlo methods are particularly effective for simulating credit loss distribution and for evaluating tail risk measures, and they are computationally less intensive than analytical methods.

The distribution of portfolio credit risk is highly skewed and has a long fat tail. Unlike the case for a normally distributed loss distribution, knowledge of the first two moments of the credit loss distribution provides little information about tail risk. To compute tail risk (large losses that occur with a low probability) one has to simulate the credit loss distribution using Monte Carlo techniques. In this entry we will provide a brief introduction to Monte Carlo methods and subsequently describe the computational process involved in performing a Monte Carlo simulation to generate the distribution of credit losses. Simulating the credit loss distribution is discussed under ...

Get Encyclopedia of Financial Models, 3 Volume Set now with the O’Reilly learning platform.

O’Reilly members experience books, live events, courses curated by job role, and more from O’Reilly and nearly 200 top publishers.