Zigzag indicator and neural nets - page 10

 
Mathemat:
Piligrimm:

For anyone interested, I can share the "formula for happiness"!

GP1[iq+1] = 
There you are, you reindeer...

Don't you understand the question?
 
I didn't ask any questions, and there were no questions from your side (if you mind - then from yours) either. I just remembered the classic bearded anecdote, that's all.

As far as I understand it, it's a polynomial from the preceding four values (of what, I don't know; here - apparently, the ZZ fractures). I'm not looking for a formula for happiness in the form of a fourth-degree polynomial; if you've found it and expect to get a TC with pf~25 using it, that's really a real super. I just wish you all the best - and please don't think I'm being snide here...
 
Mathemat:
I didn't ask any questions, and there were no questions from your (if you're against it - then from yours) side either. I just brought up the classic bearded anecdote, that's all.

As far as I understand it, it's a polynomial from the preceding four values (of what, I don't know; here - apparently, the ZZ fractures). I'm not looking for a formula for happiness in the form of a fourth-degree polynomial; if you've found it and expect to get a TC with pf~25 using it, that's really a real super. I just wish you all the best - and please don't think I'm being snide here...

I called this polynomial "formula of happiness" as a joke, that is why I put it in quotes (I specify, so that no one is under an illusion that happiness can be achieved by a single formula). Although this formula really works in my program, as well as that which I, as a "wild", gave further, and quite without jokes I propose to use it, although you should not count on the result that I have, since it is torn from the context of the whole program and trained for certain conditions, but nevertheless you can try to use it. And I think it is quite possible to reach profit 25, although certainly not with a single formula. The solution of this problem requires, firstly, an effective strategy and, secondly, a powerful expert system that analyses hundreds of parameters and works in real time.

The main thing I wanted to say with all my posts in this thread is that both Zig-Zag and neural networks can be used quite effectively, and one can achieve quite decent Forex forecasting with their help as well as by other methods, but first of all, you need a systematic approach to the solution of this problem.

It is necessary to accumulate a great potential in terms of strategies and modeling, and it may take years of hard work, there are no easy ways, and you should have no illusions in this respect, otherwise your time will only be wasted.

 
Ivikus:
SK. wrote (a):
Ivikus wrote (a):
Zigzag will always be there. It's just that sometimes you can see it and sometimes you can't. :)


You will see the zigzag, but it will be useless. In fact, a zigzag is just a coarsening of initial data with a lag. There is nothing more.

Alex-Bugalter wrote (a):
You probably just don't know how or don't want to cook them:)

I could cook it, only it's inedible:)

Not coarsening, but eliminating noise. What is noise is for everyone to decide for themselves. Not all zigzags are delayed.

Are there any examples of non-lagged zigzags?
 
AndyGri:
Ivikus:
SK. wrote (a):
Ivikus wrote (a):
Zigzag will always be there. It's just that sometimes you can see it and sometimes you can't. :)


You will see the zigzag, but it will be useless. In fact, a zigzag is just a coarsening of initial data with a lag. There is nothing more.

Alex-Bugalter wrote (a):
You probably just don't know how or don't want to cook them:)

I could cook it, only it's inedible:)

Not coarsening, but eliminating noise. What is noise is for everyone to decide for themselves. Not all zigzags are delayed.

Are there any examples of non-lagged zigzags?
With expert systems giving predictions.
 
Piligrimm, I read you carefully. But I'm sorry if I'm dumb, I still don't get it. Do you have a network with only zigzagging??? From what I've read, I understand that it does. And this is the circumstance that surprised me quite a bit. I myself see the zigzag as an additional input of a very old and highly coarse history, that's all. And of course, as a possible (though IMHO not quite optimal) prediction target. Tell me if I understand you correctly ? Again, I apologize if the question is silly, I'm just a bit surprised by this approach.
 

Hello all

I have achieved using Zigzag Swing_ZZ with period 4 on Eurobucks h1 with Fibo ratio

Indicator and Expert Advisor in the archive

Files:
 
olyakish:

Hello all

I have achieved using Zigzag Swing_ZZ with period 4 on Eurobucks h1 with Fibo ratio

Indicator and Expert Advisor in the archive


Just cool. I take it this is a test on a demo account, not in a tester?
 

I think I'm not mistaken if I say that this is a test in the tester, and in the optimization section...

Author will correct me if I'm wrong :-)

 
Neutron:

I think I'm not mistaken if I say that this is a test in the tester, and in the optimization section...

The author will correct me if I'm wrong :-)

You are not wrong :)

it really is a tester

it's true i have Eurobucks history (minutes) till 2007 04 12

and the test (chart) is for the period 20040617-20070412 ( almost three years)

it is possible to download the latest data on this pair and run

if you do not mind to put a picture how it behaves on history (more 2007 04 12)

will be grateful

I use EDGE and I have to pay for the traffic :(


ZZZY parameters as you claim to optimize is exactly the levels 23.6; 38.2; 61.8; 76.4 ;