Yoghurt systems and canned systems or The relationship between trading tactics and the reliability of historical test results - page 13
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
Let's all try to find a consistent pattern together, I just wonder if all those discussing this topic will come out of this good? Together we will try to determine the optimal window for forward testing, its suitability, etc. For starters, what should we use as a signal for entry?
An attempt to find one of the (I would say basic) regularities has yielded some result and the necessary tools for the inquisitive to continue exploring. The fruit of constructive collaboration can be seen.
Laws are working.
It is certainly interesting to look for new ones, but it is unlikely to ignite people's enthusiasm.
1 - Deviation of the signal line from the histogram. signals for Buy and Sell.
2 - Peak on the histogram. Peak at Buy may be above zero and may be below zero, the same goes for Sell, total 4 signals; I think this may help with classification.
3 - Intersection with zero on the histogram.
4 - Take into account the histogram peak and the area up to the intersection with zero. The area should be larger than a threshold value, or may lie within a certain interval - you can find that out later. The area is a kind of oversold level
These are the basic signals. If you have any other sensible ideas suggest them. Or let us start critiquing these signals and select the one we believe to be the best
These are all regularities that can be seen by the human eye and are available for human understanding. But there are patterns that cannot be detected and understood by man himself. So, they can be found only by experimenting with the optimizer and indicators, including the neural network. These patterns are the most stable in my opinion.
These are all patterns that are visible to the human eye and accessible to human understanding. But there are patterns that cannot be detected and understood by man himself. That is, they can be found only by experimenting with the optimizer and indicators, including neural networks. These are the most stable patterns, in my opinion.
Maybe, it is better to work with the laws that the whole world can see, and not only the NS?
Although I ask a question, maybe the whole world just uses NS and laughs at such simple means...
By the way, have you ever tried to teach the multiplication table to NS? :))
Just mix up the sampling first...
Maybe it is better to work according to the patterns that the whole world sees, not just the NS?
By the way, have you ever tried to teach the multiplication table to the NS? :))
Just mix up the sampling first...
In my humble opinion, working on patterns that the whole world sees is a sure loser. Although I can not say for sure - it depends on how you work.
And why teach a neural network the multiplication table? Will someone pay me for it? Not really. I am sorry, I cannot draw an analogy between Forex and multiplication table. Then why would I teach a neural network the multiplication table?
In my humble opinion, working according to the patterns the whole world sees is a sure loser. Although I can't say for sure - it depends on how you work.
And why teach the multiplication table to a neural network? Will someone pay me for it? Not really. I am sorry, I cannot draw an analogy between Forex and multiplication table. Then why should I teach multiplication table to a neural network?
There is a pattern in multiplication table :), and every schoolboy knows it.
But in forex, it is questionable, but nevertheless, in theory, the network should easily find a solution... I emphasize "should"...
There is a pattern in the multiplication table :), and every schoolboy knows it.
But in forex, it is questionable, but nevertheless, the network should be able to find a solution easily... I emphasize "should"...
Well, a pattern is not the same as a pattern. Multiplication table and forex are two fundamentally different things. They cannot be linked in any way. What conclusions can be made about Forex, if you try to teach the network the multiplication table?
I quote the person who suggested it to me: Try it - this is to see how accurately the network can predict multiple choices.....
No conclusions useful for forex, but as for NS, you can. Yes I'm not forcing you to train the network in multiplication, I just suggested you try it.
To quote a person who has advised it to me: Try it - this is to see how accurately the network can predict multiple choices.....
No conclusions useful for forex, but as for NS, you can. I do not force you to train the network in multiplication, I just offered to try.
I understand you. I've highlighted the main idea in red. For the sake of experiment and interest it may be possible. But I am not interested in experimentation for the sake of experiment, so I don't see any sense in teaching multiplication tables to a net. Accordingly, I recommend you to read carefully my first post, as it was written not for the sake of experiment, but based on my modest experience.....
Otherwise, suppose there is a pattern that appears once a month and only on higher timeframes.
But it will not be enough for skilled traders. Let them enter the market every day and every minute ... And that's why the regularity does not work.
The same MACD and stochastic were created for D1, ((((() in those days there was simply no information available, but only through reading an economical bulletin.
I.e. MACD and stochastic were not sharpened for periods shorter than 3 hours.
If I see patterns on H4 but cannot find them on H1 why do I need MTC?
===!!! MTS force me to work without any regularities! ===
We have a gap, or rather a gulf in timeframes between trading technique and MTS.
That's what I'm waiting for from you. Which signal shall we investigate? I'm more inclined towards MACD...
It would no longer be a general pattern, but some specific indicator in relation to given conditions (e.g. entry/exit). Something similar, but with Stochastic I suggested see at the end of the page.