Re: Business risk models

From: Antoine GOMIS (Antoine.GOMIS@poleia.lip6.fr)
Date: Wed Feb 06 2002 - 14:47:58 MET

  • Next message: ISCL 2002 Organization: "Summer School on Computational Logic"

    VaR (Value-at-Risk) approaches-based-business risk models and management have
    been very popular.Unfortunately , for technical and theoretical reasons, they
    fall short on catching critical events with low-probability of occurrence or
    phenomena under extreme uncertainty.These "matters of concerns" are precisely
    the core business of Risk Managers ! Today, from my personal view, there are
    three fast speed momentum gaining approaches for business Risk modelling:

     1- IPMU(Information Processsing and Management under Uncertainty) from AI-like
    blend of Artificial Neural Networks, Genetic Algorithms,Fuzzy Logic, Formal
    Concept Analysis, Rough Sets and more. I may recommend e.g Sophia Langley's
    paper "Representations of Risk" in Proceedings Seventh International Conference
     IPMU,july 6-10,1998,Paris, la Sorbonne,pp.1068-1075.Several other valuable
    models of uncertainty processing are inspirational for business risk modelling
    (e.g Bouchon-Meunier, Akdag, Grabisch, Jaffray, Mellouli,...).
     
     2- EVT (EXTREME VALUE THEORY) (Reznick, Davis, Deheuvels, Kluppelberg,
    Rootzen,...):
     This foundational theory deals precisely with phenomena under extreme
    uncertainty/scarcity with for example a potential highly "destructive"
    outcome/impact. It shows for example that under appropriate conditions:
          -number one: the distribution of extrema series converges asympotically to
           the basic Frechet, Weitbull or Gumbel distributions.
          -number two:the excess distribution over a given threshold converges to a
           generalized Pareto distribution.
      I strongly recommend the scientific journal " EXTREMES" to harness the power
       and the limitations of this theory.
          
     3- GIONICS ( a computational Intelligence approach to Uncertainty) (work under
    completion progress). A Gion is a mathematical complex (non-necessarily real)
    solution to the following equation:
      
                          2
                         X == X + 1 (Anais-Camilia-Aleea's gionic equation)
                         
                         
    The derived mathematical ( algebraic, geometric and topological ) properties and
    structures of the gion are intended to bring radical new fundamental scientific,
    technological and metaphysical insights and foresights .They maybe the missing
    mathematical framework of several "principles and postulates" in Mathematics (
    e.g "the local change of global properties, structures, functionalities" to be
    coined "transpolation"), Physics (a geometrical account of the Heisenberg's
    undeterminacy) and in other fields.As a matter of fact, it seems to be the
    missing mathematical foundation of the resounding and visionary DUKAS General
    Theory of Evolution of any variable parameter ( financial, economic, social,...)
    to be represented as the motion of a corresponding wave/particle ! Please visit
    WWW.dukascopy.com
    with a link to the Technical analysis sites ranking reported in the 06/01/2002
    Financial Times. Such an approach is being put to work for business risk model
    building.

    antoine
           
    >X-pt: isis.lip6.fr
    >Date: Wed, 30 Jan 2002 19:41:20 +0100 (MET)
    >Originator: fuzzy-mail@dbai.tuwien.ac.at
    >From: "Harris Georgiou" <xgeorgiou@yahoo.com>
    >To: Multiple recipients of list <fuzzy-mail@dbai.tuwien.ac.at>
    >Subject: Re: Business risk models
    >X-Listprocessor-Version: 6.0c -- ListProcessor by Anastasios Kotsikonas
    >X-Comment: Fuzzy Distribution List
    >
    >"Dave" <dad717@hotmail.com> wrote in message
    >news:c18ffa8d.0201291636.207e92e6@posting.google.com...
    >> Hello group,
    >>
    >> Could someone point me in the right direction.
    >> Is fuzzy logic still dominent for business
    >> risk or is another model more used today
    >> (i.e. neural?)
    >>
    >> Who is the best in programing these models?
    >>
    >> TIA
    >
    >Actually, traditional risk management is based rather on analytical
    >approaches like Game Theory rather than AI heuristics. However, any model
    >that is constructed to deal with gain-oriented goals (minimum loss vs
    >maximum gain), instead of simple success rate, can be easily applied in
    >economics and business management. As the problem becomes more complex and
    >includes more contractions in evidence (data), fuzzy rules/sets and
    >non-linear clustering (like ANN) are the best choices.
    >
    >
    >
    >--
    >
    >Harris
    >
    >- 'Malo e lelei ki he pongipongi!'
    >
    >
    >
    >
    >############################################################################
    >This message was posted through the fuzzy mailing list.
    >(1) To subscribe to this mailing list, send a message body of
    >"SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
    >(2) To unsubscribe from this mailing list, send a message body of
    >"UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
    >to listproc@dbai.tuwien.ac.at
    >(3) To reach the human who maintains the list, send mail to
    >fuzzy-owner@dbai.tuwien.ac.at
    >(4) WWW access and other information on Fuzzy Sets and Logic see
    >http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
    >(5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html
    >

    Zero is non null !

    ############################################################################
    This message was posted through the fuzzy mailing list.
    (1) To subscribe to this mailing list, send a message body of
    "SUB FUZZY-MAIL myFirstName mySurname" to listproc@dbai.tuwien.ac.at
    (2) To unsubscribe from this mailing list, send a message body of
    "UNSUB FUZZY-MAIL" or "UNSUB FUZZY-MAIL yoursubscription@email.address.com"
    to listproc@dbai.tuwien.ac.at
    (3) To reach the human who maintains the list, send mail to
    fuzzy-owner@dbai.tuwien.ac.at
    (4) WWW access and other information on Fuzzy Sets and Logic see
    http://www.dbai.tuwien.ac.at/ftp/mlowner/fuzzy-mail.info
    (5) WWW archive: http://www.dbai.tuwien.ac.at/marchives/fuzzy-mail/index.html



    This archive was generated by hypermail 2b30 : Wed Feb 06 2002 - 14:52:00 MET