Should Risk Managers Rely on Maximum Likelihood Estimation Method While Quantifying Operational Risk?

23 Pages Posted: 18 Dec 2007 Last revised: 22 Jun 2016

Date Written: February 18, 2008

Abstract

The paper compares the performance of four estimation methods, including the maximum likelihood estimation, which can be used in fitting operational risk models to historically available loss data. The other competing methods are based on minimizing different types of measure of the distance between empirical and fitting loss distributions. These measures are the Cramer-von Mises statistic, the Anderson-Darling statistic, and a measure of the distance between the quantiles of empirical and fitting distributions. We call the last method the quantile distance method. Our simulation exercise shows that the quantile distance method is superior to the other three methods especially when loss data sets are relatively small and/or the fitting model is misspecified.

Keywords: Operational risk, loss distribution approach, log-t distribution, maximum likelihood estimation, Cramer-von Mises statistic, Anderson-Darling statistic, quantile distance, simulated annealing

JEL Classification: C13, C15, G28

Suggested Citation

Ergashev, Bakhodir, Should Risk Managers Rely on Maximum Likelihood Estimation Method While Quantifying Operational Risk? (February 18, 2008). Available at SSRN: https://ssrn.com/abstract=1075249 or http://dx.doi.org/10.2139/ssrn.1075249

Do you have negative results from your research you’d like to share?

Paper statistics

Downloads
551
Abstract Views
2,923
Rank
92,247
PlumX Metrics