Online Program Home
  My Program

Abstract Details

Activity Number: 492 - Extreme Value Models and Data Visualization of Catastrophic Risk Analysis
Type: Invited
Date/Time: Wednesday, August 2, 2017 : 10:30 AM to 12:20 PM
Sponsor: Casualty Actuarial Society
Abstract #322235 View Presentation
Title: Extreme Value Models and Data Visualization of Catastrophic Risk Analysis
Author(s): John Elbl* and Bob Frady* and Steve Kolk* and Danny Stout*
Companies: AIR Worldwide and Hazard Hub and Independent Consultant, Casualty Actuarial Society member and Statistica Advanced Analytics
Keywords: Extreme Value Models ; Data Visualization ; Catastrophic Risk Analysis ; Hurricane Storm Surge ; Flooding ; Communicating Risk Statistics
Abstract:

This panel will present both proven and emerging extreme value models and data visualization. They will illustrate models that quantify extreme risks and show how to communicate model findings effectively and balance short-term insurance perspectives with longer-term community risks as follows: John Elbl, risk scientist with AIR Worldwide, will present their 'real world'-based hazard models that simulate the impacts of near-term hurricane storm surge and inland flooding (both on and off of flood plains) for the U.S. to facilitate Catastrophic Risk Analysis. Bob Frady, CEO of Hazard Hub, will speak about extreme event hazard data. He'll talk about how essential it is to rely on carefully curated, scientific data when assessing extreme event risks. He will illustrate this by using Hurricane Matthew residential damages to show how scientific Surge Risk data can add lift (=additional explanatory power) to insurance pricing of hurricane storm damages. Steve Kolk, member of the Casualty Actuarial Society, will present a new historical extreme climate data tool, the Actuaries' Climate Index. Dev Kannibaran, Sr data miner at Dell Statistica, will illustrate how extreme value models can be constructed from the sea level rise component index of the Actuaries' Climate Index.


Authors who are presenting talks have a * after their name.

Back to the full JSM 2017 program

 
 
Copyright © American Statistical Association