Predictability of Extreme Events and Its Automotive Applications

Paper #:
  • 2018-01-0094

Published:
  • 2018-04-03
Abstract:
Many engineering, natural, human, and social phenomena are dominated by extreme events, which might not always be conveniently characterized by the extreme value theory. There are methods such as the generalized Pareto and extreme distribution function theories are derived from the extreme value theory that can be used to describe most extreme events. However, recent researches show that the extreme value theories do not work well for some extreme events especially at the tails of the events, which can be rather well described by the Dragon Kings theories. Furthermore, there are other extreme cases which intrinsically lay at the very tails of the events in the regions of <<1e-6 probabilities. These extreme events, might be defined in practices as virtually unpredictable and simply known as 'Black Swan' events. This paper provides a brief introduction to the extreme events theories and the associated characterization approaches giving examples from a wide variety of cases such as warranty returns, rainfalls, terrorist attacks, city population, and earthquakes. The methods are considered in the light of developing a commonality in characterizing extreme 'Black Swan' events in general. Further work is in progress to identify the normalization and quantification of 'Black Swan' data without regard to the actual differences or causes of the occurrence
Access
Now
SAE MOBILUS Subscriber? You may already have access.
Buy
Attention: This item is not yet published. Pre-Order to be notified, via email, when it becomes available.
Select
Price
List
Download
$22.00
Mail
$22.00
Members save up to 36% off list price.
Share
HTML for Linking to Page
Page URL

Related Items

Event
2018-04-10
Technical Paper / Journal Article
2011-05-17
Article
2017-07-26
Training / Education
2007-03-01