Re: Intelligent Trading Systems

Michael D. Hofer (featheredfrog@geocities.com)
Wed, 21 Oct 1998 12:28:01 +0200 (MET DST)

Martin Sewell wrote:
>
> Hello Everyone
>
> I am about to embark on a PhD in Intelligent Trading Systems, but am being
> held back due to lack of a definite sponsor. I have a degree in Maths, a
> degree in Computing and experience of FX and commodity trading and am
> desperate to find a bank to sponsor me. It would be, among other things, an
> extremely cost effective way of gaining access to some of the top academics
> in this field. I am happy to submit a project proposal before any financial
> commitment is made. If you can help, or have any suggestions, I would be
> most grateful to hear from you. Below is an introduction to the topic.
>
> The three key technologies employed by intelligent systems today are neural
> networks, genetic algorithms and fuzzy logic. These are more recent
> developments than the previous generation of such systems, which relied on
> expert systems, although expert systems are often used as the shell for
> other technologies. In the last few years the most popular method was to
> build the system around neural networks. These, however, were of limited
> success. Although a useful tool for many applications, neural networks tend
> to fail if applied to financial markets because the markets are too
> 'difficult'. That is, they tend to be nearly, but not quite, random. The
> other main drawback to neural networks is the lack of explanation
> capabilities. It is difficult to explain how and why neural networks arrive
> at the results that they produce.
>
> The key technology that we will employ is the genetic algorithm (GA). As
> each technology has inherent strengths and limitations, they are often
> combined, using the object-oriented paradigm to produce an Intelligent
> Hybrid System. We will therefore employ fuzzy logic as the front end of our
> system if necessary.
>
> Although not an aspect of artificial intelligence, there is a branch of
> mathematics that has been making increasing contributions to our knowledge
> of the markets in recent years. Nonlinear dynamics has the ability to
> identify the existence of chaos in a market, and can give us an indication
> of when short-term predictability should be possible and also help us choose
> optimal parameters when designing a trading system.
>
> Thanking you in advance for any help.
>
> Martin Sewell
> London, UK
> 0181-343 4708
> martin@msewell.force9.co.uk

I don't know, but this _sounds_ like it would be right up D.E.Shaw's
alley - if they're not already doing it, in which case they might pay
you lots of money to go into some other field! [JOKE! IT's A Joke,
folks! Put that solicitor down carefully and move away from it before
somebody gets hurt.]

D.E.Shaw (in NYC) does algorithmic trading.

michael (who took a course from him when he was still Prof. Shaw!)

-- 
Cian ua'Lochan /mka/ Michael D. Hofer
I'm not a medievalist - I just play one on weekends!
http://www.geocities.com/SoHo/Lofts/9800/

############################################################################ 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