[Forum SIS] Seminario Prof. Ron Kenett 11-11-2009

[Dip. Scienze Statistiche - Mi]

  Con preghiera di massima diffusione


Prof. B. Vittorio Frosini, Head of the Statistics Department of the Catholic University of Milan, is pleased to inform you that Prof. Ron Kenett Chairman and CEO of the KPA Group<http://www.kpa-group.com/> and KPA Ltd., International Research Professor at the Center for Research in Risk Engineering<http://www.polyrisk.org/node/6>, NYU-Poly, New York, USA will take a seminar with the following title:


On Black Swans and the Taleb Quadrants
Wednesday 11th November at 2,30 p.m.
in the Room NI. 010 in Via Nirone, 15



Abstract
Important consequential risks are typically unpredictable and rare.  While predictable risks may be prevented, unpredictable risks test our resilience and our ability (agility) to respond. The financial meltdown of 2008 is an example of  such risks with an aggregate fate of close to 1000 financial institutions (including busts such as FNMA, Bear Stearns, Northern Rock, Lehman Brothers, etc.) who lost over 1 Trillion dollars on a single error, more than was ever earned in the history of banking.
To confront this evolving risk reality, N. N. Taleb suggested mapping randomness and decision making into a quadrant with two classes of randomness and decisions. The type of decisions referred to as "simple" or "binary" lead to decisions such as "very true or very false",  "matters or does not matter".  By the same token, statistical tests in the control of quality may state, "A product is fit for use or the product is defective".  Statements of the type "true" or "false" can then be stated with some confidence interval.  A second type of decisions is more complex, emphasizing both its likelihood of occurrence and its consequences.
Two layers of randomness, very distinct qualitatively and quantitatively,  are suggested by Taleb.  A first layer is based on "forecastable events", implied in finite variance (and thus thin tail probability distributions) and a second based on "unforecastable events", defined by probability distributions of fat tails. In the first domain, exceptions occur without significant consequences since they are predictable and therefore preventable. The traditional random walk, converging to Gaussian-Poisson processes provides such an example.  In the second domain, large consequential events are experienced which are more difficult to predict.  "Fractals" and infinite variance (Pareto-stable and chaotic) models provide such examples.


The attendance to the seminar is free of charge but the reservation is recommended because the places are limited.

Please confirm your presence by e-mail to: dip.scienzestatistiche at unicatt.it<mailto:dip.scienzestatistiche at unicatt.it>
For further information please contact directly the Organizational Secretariat
Mrs Barbara Villa
Phone. 02-7234.2647 - Fax 02-7234.2671
e-mail: dip.scienzestatistiche at unicatt.it



please don't print this e-mail unless you really need to




-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20091013/6f0da9a8/attachment.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: Ivy.gif
Type: image/gif
Size: 5665 bytes
Desc: Ivy.gif
URL: <http://www.stat.unipg.it/pipermail/sis/attachments/20091013/6f0da9a8/attachment.gif>