In this article, we propose a new model for the dynamics of the aggregate credit portfolio loss. The model is Markovian in two dimensions with the state variables being the total accumulated loss Lt and the stochastic default intensity λt. The dynamics of the default intensity are governed by the equation t=κ(ρ(Lt, t)−λt)dt+σ√λtdWt. The function ρ depends both on time t and accumulated loss Lt, providing sufficient freedom to calibrate the model to a generic distribution of loss.

We develop a computationally efficient method for model calibration to the market of synthetic single tranche CDOs. The method is based on the Markovian projection technique which reduces the full model to a one-step Markov chain having the same marginal distributions of loss. We show that once the intensity function of the effective Markov chain consistent with the loss distribution implied by the tranches is found, the function ρ can be recovered with a very moderate computational effort. Because our model is Markovian and has low dimensionality, it offers a convenient framework for the pricing of dynamic credit instruments, such as options on indices and tranches, by backward induction. We calibrate the model to a set of recent market quotes on CDX index tranches and apply it to the pricing of tranche options.

Authors: Andrei V. Lopatin, and Timur Misirpashaev

Download Numerix Research Paper

Complete the form below to download this complimentary research paper.

Select Form: 

Form #5: Research

Keep me informed of future research from Numerix:

Sign me up to receive "Thinking Derivatively" monthly newsletter by Numerix:

* Required fields
conference

Asia Risk Congress

product

CrossAsset - Leading the Industry in Advanced Models and Methods

content collection

Numerix Quant Tech Resource Hub

on-demand webinar

SRP Europe Conference 2021: Optimizing Financial Valuations to Improve Investor Experience

on-demand webinar

QuantMinds 2020: Modelling Energy Curves for XVA

conference

QuantMinds in Focus

on-demand webinar

Quantitative R&D Innovations Update

on-demand webinar

Neural Networks with Asymptotics Control

quantitative research

Machine Learning: Deep Asymptotics

quantitative research

Multi-curve Cheyette-style models with lower bounds on tenor basis spreads

written blog

Leading Numerix Through the Pandemic: Taking Action that Makes a Difference

next 2020 session video

Numerix Quantitative Research and Development Roundtable Discussion