SMS scnews item created by Jean Yee Hwa Yang at Fri 28 Mar 2008 1552
Type: Seminar
Distribution: World
Expiry: 4 Apr 2008
Calendar1: 4 Apr 2008
CalLoc1: Carslaw 375
Auth: jeany@c220-239-0-232.belrs4.nsw.optusnet.com.au

# Statistics Seminar: Nader Tajvidi -- Statistical modelling of Extreme Values: basic theory and applications to analysis of temporal trend in wind storm losses and temperature data

Speaker: Dr Nader Tajvidi (Centre for Mathematical Sciences, Lund Institute of
Technology, Sweden)

Date: Friday, 4 April, 2.00pm Location: Carslaw 375

Title: Statistical modelling of Extreme Values: basic theory and applications to
analysis of temporal trend in wind storm losses and temperature data

Abstract: Statistical extreme value theory provides a flexible and theoretically well
motivated approach to the study of large losses in insurance.  We give a brief review of
the modern version of this theory and a ¡¨step by step¡¨ example of how to use it in
large claims insurance.  The discussion is based on a detailed investigation of a wind
storm insurance problem.  A topic of major current interest in extreme-value analysis is
the investigation of temporal trends.  For example, the potential influence of
"greenhouse" effects may result in severe storms becoming gradually more frequent, or in
maximum temperatures gradually increasing, with time.

One approach to evaluating these possibilities is to fit, to data, a parametric model
for temporal parameter variation, as well as a model describing the marginal
distribution of data at any given point in time.  In this talk we discuss some
parametric trend models and illustrate the methods by application to a dataset on
windstorm losses in south of Sweden.  We shall also discuss diffculties which might
arise in formulating structural trend-models.  Motivated by datasets on windstorm
severity and maximum temperature, we suggest a nonparametric ap- proach to estimating
temporal trends when fitting parametric models to extreme values from a weakly-dependent
time series.