An expert with intelligence. Concept. - page 9
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
I ran through all the threads, and as a result: nobody has understood EricGR's idea (I assume many of those who are aware of the topic have refrained from publicly developing it), although he has presented the whole point.
I agree that the market should be followed without betting on a prediction.
The more accurate the entry, the less the drawdown.
Exit is probably the most complicated mechanism in an EA with this trading method.
well, one person's opinion and support... it's not enough to make people believe it) you need to convince half the forum and prove you're right...
why not make the project public... anyway, the final parts of a good one are always hidden and many elements of the advisor's work are not divulged...
I do believe, last May I had a successful scalper on M5 - traded in 2 directions, only in profit from 20.0 to 100.0 pps, but on the flat, could only break-even within a session
Here's a trend advisor without any secrets.
It's pretty decent.
I ran through all the threads, and as a result: nobody has understood EricGR's idea (I assume many of those who are aware of the topic have refrained from publicly developing it), although he has fully laid out the gist.
I agree that the market should be followed without betting on a prediction.
The more accurate the entry, the less the drawdown.
Exit is probably the most complicated mechanism in an EA with this trading method.
I don't think anyone has anything to hide) and I don't think I've come up with anything new, it's all about properly integrating the execution of the method into the overall idea, advantages separate in one area or another will be useless.
Ran through the whole thread, as a result: no one has understood EricGR's idea (I assume many of those who are aware of the topic have refrained from publicly developing it), although he has completely laid out the gist of it.
And I ran across the thread and did not understand what the EA's intelligence will be? "Let's make the EA trend-wise", "let's make it flexible", "let's use volatile stops"... And about intelligence when? From the wiki:
Intellect is a capability that combines all the cognitive abilities of an individual: sensation, perception, memory, representation, thinking, imagination...
Intellect as a capability is usually realised with the help of other abilities. Such as: the ability to cognize, to learn, to think logically, to organize information by analyzing it, to determine its applicability (to classify), to find connections, patterns and differences in it, to associate it with similar ones, etc.
It is clear that in this case we are talking about artificial intelligence, but nevertheless, the premise of the topic-starter to create such an expert is wrong. Of course, we will never teach an EA to "think logically ", butto "learn, systematize information by analyzing it, classify, find patterns and differences, associate it with similar, etc." it is quite possible. Especially since we are only talking about a concept, not an implementation.
The concept should be as follows: we are putting some mathematical apparatus, learning and analysis capabilities, a trading module, etc. into an Expert Advisor. What should go into it and how to do it better, we can discuss within this branch.
And then the ideal picture is the following: we put the EA on the chart and indicate, let it be the profit for the year that we want to get. For example:
- 5000% per annum. EA shakes its beats and gives us the message "It is impossible", and at the same time it orders for us a place in the nuthouse or a luxurious mental hospital, depending on the size of our deposit.
- 500% per annum. The Expert Advisor thinks for a long time and says "There is a chance to make the specified profit, but it will be risky, there can be significant drawdowns up to 50%. Are you sure?" Next we think.
- 100% per annum. The Expert Advisor thinks it over, cheerfully reports "It will be executed my master" and gets down to business.
This is ideal, but it works for the end. In principle, almost everything that is required for such an advisor, I have even realized, though not within the framework of a single program. So, nothing is impossible.
... We will never teach the Expert Advisor to "think logically", butto "learn, systematize information by analyzing it, classify it, find patterns and differences, associate it with similar ones, etc." is quite possible ...
I don't think anyone has anything to hide) and I don't think I've come up with anything new, it's all about properly integrating the execution of the method into the overall idea, the advantages of separate in one area or another would be useless.
An interesting concept of a self-adapting Expert Advisor to the market based on the mutual correlation of currency pair movements (continuing, forming a joint index of currencies based on their correlation in the range of 0.7 to 1 and making trading decisions) is outlined from this page of the magazine...
EricGR:
не думаю что у кого то есть что то скрывать) и не думаю что я придумал что то новое, все дело в правильно интеграции исполнения метода в общую идею, преимущества раздельные в той или иной области бесполезны будут.
You have to decide whether you are hiding something or not.
I'm also careful not to say anything unnecessary at this stage... At this stage the entries are at the lowest TF allowed by the trading system rules and the exit is higher depending on the circumstances... ok with this debate... everything has to be done by yourself... ... or you can show the whole thing and then listen to the criticism of an enlightened one... then it will be more objective.
Figar0:
This thread should not be taken seriously, as the answers to such questions should be found independently. Personally, I tried to understand what the topicstarter wants, and came to the conclusion that he is looking for a solution to another problem (different from the topic), I could be wrong.
Regarding adaptation, self-learning, intelligence: I abandoned this topic, because I found a system capable of showing the best results on the whole story with static input parameters.