Re: AI - WHY? Trading Systems predicting indices.

Robin B. Lake (rbl@hal.EPBI.CWRU.Edu)
Tue, 22 Dec 1998 00:22:46 +0100 (MET)

Well, we can do 8 out of 10 trades at even or better using a Fuzzy-based
approach, so why would they not be able to? Trading small amounts,
10% per month after commissions is what we target.

See:
http:/apk.net/~lake

P.S. Our records since November 1 are all real-money trades, auditable
at our broker, eDreyfus.

>
> A look at the site reveals they are predicting the daily absolute
> level of the index, thus a naive forecast would probably be more
> accurate.
>
We predict whether or not the DJIA will move 100 points or more; three
stocks that will risk and three that will fall between 9:40 and 9:55 AM;
and three stocks that will rise and three that will fall at 10:30/10:15 AM.

We also produce charts that provide fuzzy matches to the next-day's
market movements.

> On the other hand, in the text on the site, they also claimed
> "Generally, the direction and extent of market movement are
> correct more than 90% of the time."
>
> FWIW, if they are 90% accurate on direction alone, they should be
> among the richest folks on the planet in no time at all. Like
> already. And I'd say the odds of that are < 1%.
>

It isn't that easy to make a profit even when you know key aspects
and features of the markets.

> ;)
>
>
> steve & eugene wrote:
> >
> > We have a fuzzy logic / neural net based system which can predict the Dow
> > Jones Industrial Average and other indices down to > 1% accuracy reliably &
> > validly - it is done every day on www. moneyweb.co.za - look at the section
> > which says expert opinion and then market forecasts.
> >
> > Prediction on individual counters is done more accurately. We also have a
> > realtime lead lag indicator done in dataengine (www.mitgmbh.de)
> >
>
Robin B. Lake, Ph.D.
rbl@po.cwru.edu
>

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