|
PureBytes Links
Trading Reference Links
|
TL wrote:
>
> >
> > Having been an avid reader of the many posts here on systems as well as
> >
> > a fulltime trader myself I have some questions and comments.
> >
> > Most everything one needs to begin system development, as well as a very
> >
> > good system appears in my favorite book "Market Wizards" .
> >
> > 1. Paul Tudor Jones--we have tested all the systems under the sun and
> >
> > found only one that works, it is "trend following" and based on a sudden
> >
> > range expansion, in a market that has been trading narrowly.
> >
> > Question: how can you test sudden range expansion, on TS, how would you
> >
> > program in "find a market trading narrowly", i.e.what would define a
> >
> > market that has been trading narrowly?
There are many ways to program this in TS, as another post suggested,
Street Smarts speaks of this.
> >
> > 2. Here is a great system that every good trader will agree works- When
> >
> > the market does not react to the news the way it should go with that
> >
> > move strongly. Every "market wizard" followed this "system". How could
> >
> > that ever be checked with a TS type program? Or programmed as a system?
> >
> > 3. How is anyone who ever looks at a bar chart able realize that one
> >
> > day's action might have been more news driven than another. Or that a
> >
> > sudden rumor triggered a selloff that ended when the rumor was denied?
> >
> > How do you factor in news? Shouldn't all systems focus on the news as a
> >
> > primary indicator?
Maybe, but of course you cannot program news analysis into a system.
Why is everyone always looking for an easy way to completely systematize
everything. I guess it is the dream of the holy grail black box that
makes money while you play golf.
Your brain is a pretty good analyzer, though. That is how the Market
Wizards do it. I don't think anyone in that book trades a mechanical
system. As we spoke here about recently, Paul Tudor Jones spent several
years and several million to "program" his brain into a neural net
trading computer called Ranger. The plan was to sell it for $50,000.
They gave up a few years ago having gotten nowhere.
> >
> > 4. Professional traders seem focused on intraday data and divergences
> >
> > between indicators, how does one program in indicator divergence? They
> >
> > also use the confirmation of other related markets, how can that be
> >
> > system tested?
> >
> > The best traders I know synthesize information from a variety of
> >
> > sources.
Again the best hardware to use to synthesize all this info is your
brain, but if you have all this info coming in to TS on a RT basis you
can program TS to do all the checking you want to find any situation you
want. Sounds like you need to spend a little quality time with the Easy
Language manual, or hire a professional TS programmer from Futures mag
ads.
> >
> > 5. Since moving averages are in the long run profitable is it just they
> >
> > are too boring for people to use or comment on? There is not a study
> >
> > out there that seems to suggest the longterm use of moving averages
> >
> > loses money. To quote Marty Schwatz "using moving averages works better
> >
> > than any other tool I have".
That is true, I use Moving averages. THey aren't sexy though, and they
aren't perfect, so many people feel there must be something better out
there. You know the attitude, if it is simple it must not work
anymore. I have to find something that no one has EVER thought of
before to defeat the market. All those market wizards say KISS, the
more rules the worse the system. They haven't been proven wrong yet.
I think this last question just made the previous questions moot. You
now have the basis for a good trading system that is completly
mechanical and preprogrammed into TS and simple.
Good Luck,
Eric
|