Recognising images ( rhetorical theme ) - page 10
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
Vadim, what do you mean by 'expert system', what can it do?
The logic of an ordinary EA consists of three parts: a signal, filters, and an executive. An ordinary EA sings what it sees: there is a signal, we pass it through filters and, if it has passed, we open an order in a certain direction. Thus, it can react to 1/1000 of the available market information. It's good if you know the dependency that works clearly, then you're kind of in the sweet spot. But somehow I haven't managed to make a simple and working Expert Advisor, so I've gone the extensive way. The expert system gathers information that characterises the state of the market and decides on the basis of this information to apply one strategy or another. I have a powerful expert and executive parts. It is a kind of accelerator for Expert Advisors. You can accelerate your average Expert Advisor to a well profitable level.
I have a powerful expert and executive part. It is a kind of accelerator for EAs. You can accelerate mid-level EAs to a well profitable level.
Ok. Vadim, if it is not too difficult and interesting, calculate the following scheme.
There is an impulse and a 3-wave correction. Such corrections often look like almost regular zigzags and after the second wave is formed( the red dot in the drawing), we can already estimate approximately the end of the 3rd wave, i.e. essentially the expected end of the correction.
Based on that, we try to jump into the trend at the expected bottom of the correction.
If the correction is 50% - 1 figure: buy, stop at 75%, profit taking into account the expected trend momentum, approx. -50%.
If correction is about 1/3, i.e. 33% - figure 2: buy, stop at 66%, profit also approx. -50%.
Well, all the rest - also participate)).
Thanks in advance.
ok. Vadim, if it's not difficult and interesting, calculate a circuit like this...
There is an impulse and a 3-wave correction. Such corrections often look like almost regular zigzags, and after the formation of the second wave( red dot in the picture) you can already roughly calculate the end of the 3rd wave, i.e. essentially the expected end of the correction.
If it is within a day, such an algorithm gives about 50/50 probability of winning, the analysis of the instantaneous state does not give a valid statistic advantage.
If we filter out the signal, say, we trade only in the direction of the global trend, it will give insignificant stat advantage. It will not yield much profit but it will not work taking into account realities of brokerage companies.
I cannot calculate this scheme, I can try to use a ready algorithm for recognizing this particular pattern to work out a profitable trading strategy. And it is most likely not as simple as in the proposed option.
I can't calculate this scheme, I can try to use the algorithm for recognizing this particular pattern to work out a profitable trading strategy. And it's most likely not as simple as the one on offer.
Give me a ready-made algorithm and the essence of trading... and then what will be left to find out from what cannot be seen in the tester...?
I just don't have time to deal with the algorithm right now, although I can imagine a couple of methods,
but it sounds like a fun idea, if there's someone free and full of creativity, we can try it...
Give you a ready-made algorithm and the essence of trading... and then what will be left to find out from what cannot be seen in the tester...?
Such an answer, as if I am asking for something ready-made and profitable :) If it were as simple as you think it is...
The essence of trading is exactly what you don't have. "A ready-made algorithm" in this case doesn't exist either. What you have in the public domain does not recognise well is one, two is that you have not researched your creations in application to the market or researched very superficially.
And I and some others in this thread are trying to explain to you that normal pattern recognition is not even one tenth of a success. If it was, you would have been trading successfully long ago by now.
I just don't have time for the algorithm right now, although I do have a couple of methods,
but it sounds like a fun idea, if there's someone free and full of creativity, we can try...
Mm-hmm. I'm not serious.
denis_orlov:
ok. Vadim, if it's not difficult and interesting, calculate the scheme...
I looked, the results are bad, when choosing parameters of swing and stops by Fibo there is a strong fitting effect, I will not even show the report, but if we apply simple stop in pips and usual trawl then the results are better (only buy, lot by volatility in the report)
Strategy Tester: serfer_modPerc_0.4.3.1(!)Also.
the results are so-so, because if Sell is added, the performance is likely to deteriorate. Optimization was done for 2008
Does the stop work correctly when testing at opening prices?
Well the log is kind of clean...
On ticks it will be the same, almost
Does the stop work correctly when testing at opening prices?
The stop is correct, but the trawl is not.
Stop is correct, but the trawl is not.
As far as I remember, when testing by opening prices, if the stop loss and take profit are in the channel of the bar formed, the position will always close at the stop loss.
The entire Expert Advisor works only on open prices, including the trawl. Perhaps if it worked differently, the trawl might not work correctly, but in my case it works like clockwork, if I'm wrong, please correct.