Univariate and Multivariate Extremes for the Environmental Sciences

Authored by: Daniel Cooley , Brett D. Hunter , Richard L. Smith

Handbook of Environmental and Ecological Statistics

Print publication date:  September  2017
Online publication date:  January  2019

Print ISBN: 9781498752022
eBook ISBN: 9781315152509
Adobe ISBN:


 Download Chapter



The underlying philosophy for the statistical analysis of extreme values is to “let the tail speak for itself.” In practice, this is often done by analyzing only data that are considered to be extreme. Additionally, the goal of an extreme value (EV) analysis is often extrapolation; one may have data with a record length of 50 years, but need to estimate the magnitude of an event which has an annual exceedance probability (AEP) of .01; that is the so-called ‘100-year event’. Because extreme events are rare, EV methods generally result in large uncertainties being associated with estimated quantities.

Search for more...
Back to top

Use of cookies on this website

We are using cookies to provide statistics that help us give you the best experience of our site. You can find out more in our Privacy Policy. By continuing to use the site you are agreeing to our use of cookies.