Forecasting Extreme Financial Risk

by Kay Giesecke of Cornell University, and
Lisa Goldberg of MSCI Barra

April 11, 2005

Abstract: Extreme value statistics provides a practical, flexible, mathematically elegant framework in which to develop financial risk management tools that are consistent with empirical data. In this introductory survey, we discuss some of the basic tools including power law distributions, the peaks over thresholds estimation procedure and point processes.

Keywords: extreme events, normal distribution, extreme value distribution, power law, Pareto distribution, peaks over thresholds, tail index, shortfall risk, Hurst exponent, clustering, contagion, point process, Poisson process, fitness testing.

Published in: ICFAI Journal of Financial Risk Management, Vol. 3, No. 4, (December 2005), pp. 53-67.

This paper is republished as Ch.23 in...

Books Referenced in this paper:  (what is this?)

Download preprint (375K PDF) 22 pages