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

RE: ZigZag validity



PureBytes Links

Trading Reference Links

Hello Igor,

Yes you are right when you are talking about X points or X% from previous
high or low.
What you find disappointing does not refer directly to ZigZag validity but
to Zig Zag indicator's behavior itself.
In fact you are referring to a very choppy market in your example. In this
kind of market any one would feel the same disappointment with you, unless
he is after each tiny fluctuation of prices and has a system that manages to
make money out of this.
But in this case using zig zag is useless if not disastrous.
What is the purpose of zig zag? It is to filter out noise! And you are
talking about noise here. So, in this kind of market every long-term system
fails, especially if the prices are close to (or oscillate around) the
threshold.
My next objection is that you reduce validity's usage to visualization only.
I cannot understand this. If an indicator is useful for visual analysis and
the program's formula language permits its calling from within other
formulas, then why shouldn't you use it this way too? It is either a useful
indicator or not. If you decide it is useful then why not call it elsewhere?
Now, Igor, in all my messages I think I keep writing more about what zig zag
and my indicator can NOT do, than what they CAN do. I keep underlining what
one should be aware of and what one should avoid. I keep talking about
risks, about zig's unsuspecting victims, about faulty buy and sell signals,
or about additional conditions if one is to use zig with validity
effectively. Look at my simple code, the one you used in your argument:

If(ZigVal and and zigCondition and myCondition1 and ..myConditionN,
doSomething,
doSomethingElse)

As you see I am talking about ..myCondition1 and ..myConditionN.
As I have already said, I would never base any system solely on zig even if
now I can measure its validity.

In my last message sent to Metastockusers you can read more on the subject.
There, I also expressed the idea of another version of zig's validity, which
would oscillate between 1(for up trend) and -1(for down trend) with no zero
values, for those who prefer to deal only with confirmed (but also delayed)
trend indications.
Personally, I don't like this approach, since it misses every neutral phase.
However, this can be made.

Regards

Spyros
-------------------------
Igor wrote:

Hy spyro,

You say  ZigValid:=  fml("ZigZag validity") = 1 .but in an uptrend when
the close of the last bar is lower then the previous one then
ZigZagvalidity becomes 0 again and if the next bar is higher then you
have the condition 1 again and if next is again lower then you have
again 0..SO when you use this formula for a system If(ZigVal and and
zigCondition and myCondition1 and ..myConditionN, doSomething,
doSomethingElse) .in practise that would mean that you have one bar a
system next bar no system next bar system next bar no system..that I
find disappointing. If I'm wright the only thing what this ZigZag
validity does is visualise in your chart a point witch is X points or X%
from the previous high or low?...Witch is like an opposite trailingstop
( in this case a start trading) on the last previous high or low.
Greetings igor