expert based on Elliott Wave Theory - page 9

 
I laid out the basic idea of my as yet unfinished script on onyx, in the corner of nen's chief ZigZag specialist. Here is the initial link: http://onix-trade.net/forum/index.php?showtopic=88&st=135#. There then is another post in which I describe the idea in more detail. So far I haven't got to the results due to purely technical difficulties. And the clustering thing is not so easy there either.

2 Integer: It may be simpler. If you already have an implemented indicator or Expert Advisor with fuzzy algorithm for recognizing at least basic wave shapes, then you are really far advanced...
 

... A picture of the location of the projection points in the future will be posted a little later (constructed in MS Excel) ...


It would be very interesting to have a look.
 
Mathemat:

Oh NYROBA, I didn't take the trouble to look through all your posts in that huge thread. Yours alone, although there are a lot of other things that are not less interesting. Here's what I found (not counting a few, to put it mildly, not very good predictions; well, everything is clear there, because predictions are a thankless business):

All the figures regardless of their scale have the same geometrical structure, and the eye recognizes and distinguishes from the monowave flow standard and non-standard Elliott figures by training. The idea was to write a program which would recognize in time the formed figures,

Elliott shape analysis is not implemented in an autobot for a week or even a month of coding, that's for sure. Especially if the supposed coder is not familiar with EWP neither in Prechter's variant, nor in Neely's variant, nor at least in Miner's variant...

Alex Niroba 05.01.07 19:27
I'm only now beginning to understand that I'm a bit too far from writing a working indicator or advisor... :)))

Yep, that was realised in January. Has anything changed? Still under illusions about its feasibility in a month?

I already have some experience of working on the real account, unfortunately for me it is negative :(
Trading hands on the demo is different from trading on the real! I assure you that with full responsibility! :)))
If you are able to get 1,5-2 thousand per month on demo, for example, it does not mean that the same thing will be on the real account. I used to think: "What's the difference between demo and real? ...last month, I changed my mind. Psychology, my ass...

No, it's hard to fall for that kind of crap now.
So, I can trust the trade only to the artificial intelligence, as sad as it sounds AD.
I would not give a shit about the program, it does not care about craps, and since it does not care about them,
so it should trade much more efficiently than my hands :))))))

Yeah, I see a definite evolution, progress is clear. One thing I don't agree with though: robots are usually much less efficient than the same thing by hand. Simply because not everything we see with our eyes and consider obvious can be just as easily and unambiguously algorithmised.

Alex Niroba 11.01.07 19:23
I wanted to see the best result Rich got in the 1st place https://www.mql5.com/ru/users/Rich/
P.S. It's sad...

This is the illustration of the problems of automation. True, not very convincing, because Rich's system itself is clearly unstable, and he clearly stated it (92% drawdown on the tested period).

To write a really working strategy, you need to understand all the details, not to mention the main points. The principles inherent in my strategy, which I "understand and apply by hand", are very difficult to explain to a programmer in words. I have already encountered this, so it's probably faster to learn the programming language myself :))).

This was written by you on February 6, 2007. All the details are not only wave figures, but also Fibo (both by price and by time, and on several TFs at the same time), and perhaps cyclic analysis, etc. etc. And then there are, for example, intermarket correlations... Are you still hoping to explain all this to a coder who is not familiar with EWP?

I've been writing my EA for probably half a year with interruptions. And I'm not in a hurry, since many ideas need to be solved in my brain. And there is no Elliott, only Fibo. And this, you understand, is only a part of EWP...

Mathemat, as far as the programme is concerned, I have no illusions about deadlines, it's clear that a month is easy :0). God give ha a year to manage!

I have a practical understanding of how exactly the market "works", mind you, not theory, but practice, supported by whatever the results are. i.e. I can describe in words how and what I do, but how to rethink it all and how to put it into code, I still don't know. That's why I've addressed for help to knowledgeable and understanding people :).

And now think about it, a whole year to correspond by mail, no one has enough patience!

That's why I decided there's no better way than personal communication.

Sincerely yours,

Alex Niroba

 

God grant me a year to do it!

Well, now at least it's clear how much work to expect. ... Dear Integer, Yurixx, aren't you scared yet?

Alex, everything is much simpler: you chop cabbage by hand (couple of months will be enough, I hope?), and then get some mere trifle from it - say, 30 thousands of euros - and buy any quite decent programmer together with his giblets for a whole year! But there won't be any need to search through forums and explain something to those who aren't able to comprehend it. You've already lost almost a year and a half trying to find a coder, and still you're stuck. I would have done exactly that, if I had had such a perfect machine for chopping greenery...

I think that the correspondence by mail will be very useful for you. Formulating your tasks in clever words, you will be able to understand them clearly. In words and in your head it's all clear and obvious, but in writing you try to formulate them so that the coder understands you...

 
Mathemat:

God willing, a year to go!

Well, now at least it's clear how much work to expect. ... Dear Integer, Yurixx, aren't you scared yet?

Alex, everything is much simpler: you chop cabbage by hand (couple of months will be enough, I hope?), and then get some mere trifle out of chopped cabbage - say, 30 thousands of euros - and buy any quite decent programmer together with his giblets for a whole year! But there won't be any need to search through forums and explain something to those who aren't able to comprehend it. You've already lost almost a year and a half trying to find a coder, and still you're stuck. I would have done exactly that, if I had had such a perfect machine for chopping greenery...

I think that the correspondence by mail will be very useful for you. Formulating your tasks in clever words, you will be able to understand them clearly. In words and in your head, everything is clear and obvious, but in writing, you try to formulate them so that the coder understands you ...

I don't think anyone wants to take the risk... including you, don't you?!

I'm probably not the only one here who has another source of income not related to forex.

All the same, I prefer my own shirt closer to my body, so I prefer a bird in the hand...

I propose to reduce risks, i.e. not to rush headlong into the job, and not to quit your main work, but to combine

with program writing (you may do it in the evenings).

And as there will be a stable result, you can change jobs for a more interesting and profitable :).

P.S. If you can not get a strategy to the code, well, nothing can be done,

At least in financial terms, you won't lose anything.

And if you get it?! Think about it...

You can spend the rest of your life doing nothing, suntanning and drinking cocktails on the beach on your own island!!! :)))

Sincerely,

Alex Niroba

 
NYROBA:

I don't think anyone wants to take the risk... and neither do you, don't you?! I am probably not the only one who has another source of income, not related to Forex. I suggest to reduce risks, i.e. not to rush headlong into the maelstrom and not to quit my main job, but to combine it with program writing (you can do it in the evenings). And as soon as you get a stable result, you can change your job to something more interesting and profitable :)

Of course, that's exactly what you have to do! I'm not going to give it up myself until I feel my feet firmly planted on the ground. I'm going my own way, trying to implement the strategy into code on my own. But I have much more ability to formulate TOR clearly than you, Alex. Just because MQL4 is a foreign language for you, while for me it will soon become my mother tongue.

By the way: If I have enough time to spend a year working with an Expert Advisor, I may find enough time to build an initial capital manually, without big risks, trading cautiously, with stop-loss. And then... Then you can "do nothing for the rest of your life, sunbathe and drink cocktails on the beach on your own island!!! :)))"

One more thing: I'm speaking to you on a first-name basis, but I get a "you" in return. If it bothers you, I'm prepared to speak to you personally on a first-name basis only.

 
Pardon me for intruding. From my own experience, skipping a lot of details, I can say only one thing: leave it alone for a while. To make it work, you have to do very little: give the machine eyes to look at graphs and an analogue brain to make decisions in an unobvious way. Without it, trying to formalize elliotchine and make MTS, believe me, is a deliberate waste of time. However, this is my private opinion and I do not insist on it. If you want to see for yourself, go ahead and master mq4 and prepare to be greatly disappointed.
 
alexjou:
Pardon me for intruding. From my own experience, skipping a lot of details, I can say only one thing: leave this venture for a while. It would only take a tiny fraction of the time to be useful, if you could endow the machine with eyes to look at charts and an analogue brain to make unobvious decisions. Without it, trying to formalize elliotchine and make MTS, believe me, is a deliberate waste of time. However, this is my private opinion and I do not insist on it. If you want to see for yourself, go ahead, master mq4 and get ready for disappointment.


alexjou, it's difficult to formalize elotchyna, no doubt, but I'm not exactly elotchyna,

i.e. not in its pure form, or rather based on wave theory, but with my modifications... :)))

to Mathemat, I have nothing against being on a first-name basis.

 
NYROBA:

You can spend the rest of your life doing nothing, sunbathing and sipping cocktails on the beach of your own island!!! :)))

Utopia.

Those who have achieved during their life anything worthwhile, such an idea did not come to them.

And those to whom it came and was delayed, usually beyond the discussion on the forum does not go further ... :)

 
Mathemat:

God grant me a year to do it!

Well, now at least it's clear how much work to expect. ... Dear Integer, Yurixx, aren't you scared yet?

When Alex first published a year and a half ago, he told me that he has a great strategy that allowed him to turn $150 into $2 million in 2 months by hypothetical trading and manually calculating on a calculator, of course, I did not believe him. Not that he did it, but that it can be implemented in real life. I wouldn't believe even a 100 times smaller result.

However, like any sane person, I understood that Alex had a strategy. Whether it was formalizable or not, whether it worked or not, whether it was correct or not was unknown. And the only way to find out is through practice. Therefore I am waiting with interest for Alex, alone or with the help of others, to achieve this same practice.

Since I myself have been developing my own approach all this time and he, strangely enough, also deals with graphic objects, I can imagine the difficulties of that path. :-)) And for the same reason I suggested Alex should teach me to program the Expert Advisor himself. And I intended to limit the work on the Expert Advisor to development of specific blanks that would form the backbone of the future Expert Advisor in that part, whose formalization is unquestionable, and that would give him the necessary programming experience.

Somehow it seems to me that no programmer (if any) will reach the end of the work. :-))) And Alex, being a very ambitious person, won't quit his job. That's why the best thing for him is to be able to write and understand MQL by himself. Even if he won't do it himself, he'll need it for effective transfer of the results of previous programmer's work to the next one.

2 NYROBA wrote: And now think about it, a whole year of correspondence by mail, no one has enough patience!!! That's why I decided that there's no better way than personal communication.

Alex, think of a compromise. Even Moscow programmers will not work (except in the state) in your wonderful office. Everyone prefers to work at his favorite place. That's why it's better to combine: 2-3 days session in person, 2-3 months of remote work with communication by mail and ICQ, then - at the beginning of the cycle. :-)