|
PureBytes Links
Trading Reference Links
|
in my experience from a few years back has been this:
you can basically do two things with them:
1. curve fit the crap out of non linear time series and then sit back
and ponder at great stats (90% profitable not uncommon )
and paper profits. try trading it in real time - better make a
donation to the crippled children fund.
2. prune it and it turns into an overgeneralizer you then
are better off just using rule based approach( aka go
trade aberration ) it will do the same thing.
also a very common experience with nn is someone comes out
with real good nn trading results because they catch a streak or
an up trend and anything works during that time including nn.
as soon as the trend ends.nn is a toast. so they have to retrain
and the cycle go on like that till they finally ditch the thing.
all nn's are good at is what your brain is good at:
look at the chart and see how easy it is to pick tops and bottoms.
top looks like a top and bottom looks like a bottom.
well same thing goes about nn's. you can make them recognize
those with precision, but....
only in 20/20 hindsight. nn are not better off predicting the those
in real time than your brain is... try picking tops and bottoms
in real time... same goes for nns.
as far as building price prediction models - price is mainly
unpredictable... it's too dynamic and what people end up
doing is generate predicted series and if price series deviates
away from the predicted ( error )- you got something's wrong and
that means you want to buy or sell at this point in time -
the approach that i do not like...
input number limitation ( over generalization vs curvefit ), input
selection.
training process, black box ( rules are unknown ), etc. are the soft
spots.
imho, basically it's a good tool if you want to paper trade forever.
i rejected nns and products like abtech as a primary model a long time ago.
best thing you can do with them is pattern recognition in hindsight...
however i wholeheartedly endorse fuzzy logic concepts. too bad there
are no clean FL trading products out there. FL is an extremely
valuable tool.
GAs are just a way to speed up optimization. no tricks in there.
would be nice if omega would include that in TS.
AI in general has great appeal to it, but it just a tool. no magic
in there.
bilo.
ps. i would be delighted if someone can show me that nns are great
tools AND are tradable. all i am seeing to this day is 90% paper profits
on past data and marginal performance on real data...
----- Original Message -----
From: Tom <profilic@xxxxxxxx>
To: pierre.orphelin <pierre.orphelin@xxxxxxxxxxxxxx>; Schedlbauer, Joseph E.
<JoeS@xxxxxxxxxxx>; Omega email list (E-mail) <omega-list@xxxxxxxxxx>
Sent: Wednesday, August 09, 2000 1:32 PM
Subject: RE: Neural Networks for trading
> How is a neural net better than or not just the same, as optimising? From
> what I have seen over the years, a neural net trading system is very
similar
> to an optimisation in its resilience.
>
> Tom
>
> -----Original Message-----
> From: pierre.orphelin [mailto:pierre.orphelin@xxxxxxxxxxxxxx]
> Sent: Monday, August 07, 2000 12:37 PM
> To: Schedlbauer, Joseph E.; Omega email list (E-mail)
> Subject: Re: Neural Networks for trading
>
>
> You may want to check this one:
> Safir-X v 3.0 ( free evaluation version available from the web site
below).
> Neurofuzzy logic systems.
>
> Also:
> Neuroshell Trader ( Ward system group)
> Biocomp Profit ( Biocomp)
> Both neural networks and GA
>
> Sincerely,
>
> -Pierre Orphelin
> Neurofuzzy Logic tools for TradeStation
> Free evaluation versions and competitive upgrades available
> web: http://www.sirtrade.com
>
>
>
> ----- Message d'origine -----
> De : Schedlbauer, Joseph E. <JoeS@xxxxxxxxxxx>
> : Omega email list (E-mail) <omega-list@xxxxxxxxxx>
> Envoy : lundi 7 aot 2000 17:17
> Objet : Neural Networks for trading
>
>
> > Is anyone aware of any trading programs which use neural networks? The
> only
> > one I've found is Trading Solutions from NeuroDimension.
> >
> > I'm looking for a program specific to trading, not just a general neural
> > network.
> >
> > Thanks in advance.
> >
>
|