Comparison of certain value-at-risk estimation methods for the two-parameter Weibull loss distribution


Gebizlioglu O. L., ŞENOĞLU B., MERT KANTAR Y.

JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS, vol.235, no.11, pp.3304-3314, 2011 (SCI-Expanded) identifier identifier

  • Publication Type: Article / Article
  • Volume: 235 Issue: 11
  • Publication Date: 2011
  • Doi Number: 10.1016/j.cam.2011.01.044
  • Journal Name: JOURNAL OF COMPUTATIONAL AND APPLIED MATHEMATICS
  • Journal Indexes: Science Citation Index Expanded (SCI-EXPANDED), Scopus
  • Page Numbers: pp.3304-3314
  • Keywords: Value-at-risk, Quantiles, Weibull distribution, Monte Carlo simulation, Deficiency
  • Anadolu University Affiliated: No

Abstract

The Weibull distribution is one of the most important distributions that is utilized as a probability model for loss amounts in connection with actuarial and financial risk management problems. This paper considers the Weibull distribution and its quantiles in the context of estimation of a risk measure called Value-at-Risk (VaR). VaR is simply the maximum loss in a specified period with a pre-assigned probability level. We attempt to present certain estimation methods for VaR as a quantile of a distribution and compare these methods with respect to their deficiency (Def) values. Along this line, the results of some Monte Carlo simulations, that we have conducted for detailed investigations on the efficiency of the estimators as compared to MLE, are provided. (C) 2011 Elsevier B.V. All rights reserved.