[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

RE: Random Walk alive and well.



PureBytes Links

Trading Reference Links

Interesting, but I think your conclusions are, may I say, random :) It's the
Black Swan induction problem. No matter how many white swans you see, you still
can not conclusively say there are no black swans. In other words, no matter
how many emails our industrious friend Pierre writes about his software, you
can not conclusively say it is not, uh..., well, let's just a black swan :)

Conclusion 1: The markets could still be random since you may not have tested
on a large enough sample set of out of sample data; your test result itself was
random and thus not conclusive.

Conclusion 2:Of course! Professional gamblers do it against worse odds and have
for a few hundred years now.Without software perhaps?

Conclusion 3: I can do a hand stand for over 3 seconds! Do I work for Cirque de
Soleil (sp?)? Not in this random lifetime. Or, just because out of that
infinite typing group you have THE one monkey that happened to create Homer's
Illiad, would you give him a million dollars to write The Odyssey next?

alexander the last infinite number of monkeys

--- Pierre Orphelin <sirtrade@xxxxxxxxxx> wrote:
> FYI,I gave a conference last year in Paris where I presented, live, the
> following experience:
> 
> True random data have been bought and were used to generate random bar
> charts.
> We used the Safir-X software to find a trading system that worked with these
> data.
> Quite easy.
> 
> The system was then tested over out of sample  random data.
> It still worked, but very badly.
> 
> Then the same system was tested on intraday futures contract ( sp, dax, ftse,
> cac, nasdaq 100).
> All of the results were very good.
> 
> The conclusion were:
> 1- The market are not random ( or we would have got the same poor results
> than
> with unseen random data)
> 2- It is possible to build winning trading systems even by learning from
> random
> data.
> 3- Safir-X was at least able to do that.
> 
> Sincerely,
> 
> Pierre Orphelin
> www.sirtrade.com
> TradeStation Technologies representative in France
> Safir-X neurofuzzy logic trading system builder
> 
> To subscribe to our FREE trading system newsletter:
> http://www.sirtrade.com/newsletter.htm
> 
> 
> 
> 
> 
> 
> > -----Message d'origine-----
> > De : Bilo Selhi [mailto:biloselhi@xxxxxxxxxxx]
> > Envoyé : samedi 9 février 2002 23:11
> > À : Rob Marsh; omega list
> > Objet : Re: Random Walk alive and well.
> >
> >
> > don't get too discouraged by the fact that
> > random walk data looks identical to
> > real market data. it does not automatically
> > mean that markets are random...
> >
> > in fact you can have a non random complex
> > non-linear process generate data that
> > looks very similar to a simple purely random
> > process... but underlying processes are totally
> > different.
> >
> > random walk is used simply because it is
> > very easy to generate and model trivial market variables
> > such as volatility, returns, etc...
> >
> > the real price generation process is  pretty complex and
> > has not been modeled as of yet ( at least not published ).
> > only in the past ten years some minor progress has been
> > made in modeling markets properly...
> > once you get there you will realize that market are
> > far from random...
> >
> > undoubtedly there is a part of the market that is random
> > there is also a part of it that is not.
> > our job as system developers is to separate the two
> > and make money on the second part.
> >
> > bilo.
> > ps. go see Beautiful Mind, it is mainly because of the
> > main character in the movie  that are able to just
> > in recent 10 years understand the markets  better...
> > that's why he got the Nobel prize almost 50 years later.
> >
> >
> >
> > ----- Original Message -----
> > From: "Rob Marsh" <dartingriver@xxxxxxxxxxxxxxxxxxxxx>
> > To: "omega list" <omega-list@xxxxxxxxxx>
> > Sent: Saturday, February 09, 2002 4:07 PM
> > Subject: Random Walk alive and well.
> >
> >
> > > While fiddling about in Excel today, quite by chance I found something
> quite
> > > sobering (from a classical chartist's point of view anyway)
> > >
> > > If you have a couple of minutes try creating the following simple Excel
> > > spreadsheet ( to generate a random walk 'price chart')
> > >
> > > In cell A2 :
> > > =rand()
> > >
> > > In Cell B2:
> > > =if(A2<=0.5,1,-1)
> > >
> > > In Cell C2:
> > > =C1+B2
> > >
> > > Copy this down (say) 3000 rows and plot a line graph of column C.
> > >
> > > Then keep regenerating the random numbers using F9 to get a freshly
> > > generated random walk each time.
> > >
> > > What do you see?
> > >
> > >     Trend lines, retracements, consolidations, breakouts, Support,
> > > Resistance, Head and            Shoulders, double tops, V-shaped
> > > bottoms...........etc....etc.
> > >
> > > Any pattern you like in fact.
> > >
> > > They are all there if you have an eye for them.
> > >
> > > AND ALL ON CHARTS THAT ACTUALLY ARE RANDOM WALK!
> > >
> > >
> > >
> > >
> > >
> >
>