Abstract
A common characteristic for many insurance risks is the right-tail risk, representing low-frequency, large-loss events. In this paper I propose a measure of the right-tail risk by defining the right-tail deviation and the right-tail index. I explain how the right-tail deviation measures the right-tail risk and compare it to traditional measures such as standard deviation, the Gini mean, and the expected policyholder deficit. The right-tail index is applied to some common parametric families of loss distributions.
Volume
2:2
Page
88-101
Year
1998
Categories
Financial and Statistical Methods
Loss Distributions
Extreme Values
Financial and Statistical Methods
Loss Distributions
Frequency
Financial and Statistical Methods
Loss Distributions
Severity
Publications
North American Actuarial Journal