Document Search

Author | |
---|---|

Title | |

Keyword | |

Abstract | |

Date: | |

To | |

Recent Comments

Value-at-Risk with Heavy-Tailed Risk Factors

**Company:**Columbia University

**Company Url:**Click here to open

**Year Of Publication:**2000

**Month Of Publication:**June

**Pages:**33

**Download Count:**1543

**View Count:**13334

**Comment Num:**0

**Language:**EN

**Source:**working paper

**Who Can Read:**Free

**Date:**9-19-2002

**Publisher:**Administrator

**Summary**

This paper develops efficient methods for computing portfolio value-at-risk (VAR) when the underlying risk factors have a heavy-tailed distribution. We focus on multivariate t distributions and some extensions thereof. We develop two methods for VAR calculation that exploit a quadratic approximation to the portfolio loss, such as the delta-gamma approximation. First, we derive the characteristic function of the quadratic approximation and then use numerical transform inversion to approximate the portfolio loss distribution. Because the quadratic approximation may not always yield accurate VAR estimates, we also develop a low variance Monte Carlo method. This method uses the quadratic approximation to guide the selection of an effective importance sampling distribution that samples risk factors so that large losses occur more often. Variance is further reduced by combining the importance sampling with stratified sampling. Numerical results indicate that large variance reductions are typically obtained. Both methods developed in this paper overcome difficulties associated with VAR calculation with heavy-tailed risk factors. The Monte Carlo method also extends to the problem of estimating CVaR.

This document is published in Mathematical Finance (volume 12, number 3) July 2002, 239-269.

http://dx.doi.org/10.1111/1467-9965.00141

This document is published in Mathematical Finance (volume 12, number 3) July 2002, 239-269.

http://dx.doi.org/10.1111/1467-9965.00141

**Author(s)**

This document's citation network:

Similar Documents: