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

详细信息    查看全文

• 作者：Omer L. Gebizlioglu^a ; ^{gebizli@ankara.edu.tr} ; Birdal Ş ; enoğ ; lu^a ; Yeliz Mert Kantar^b
• 关键词：Value-at-risk ; Quantiles ; Weibull distribution ; Monte Carlo simulation ; Deficiency
• 刊名：Journal of Computational and Applied Mathematics
• 出版年：2011
• 出版时间：1 April 2011
• 年：2011
• 卷：235
• 期：11
• 页码：3304-3314
• 全文大小：227 K

文摘

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.