Machine learning in trading: theory, models, practice and algo-trading - page 432
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
Although I would not refuse to train my stamina.....
This is the main thing I teach my students, a man is principled, not to bend, if you set out to die. This is the hardest thing in the class, I had one student (Maxim), he did not have the strength, not enough character, too feminine, shook the boy, but you will succeed you are a man!
This is the main thing I teach my students, a man is principled, not to bend, if you decide, then to death. This is the hardest part of teaching, I had one student (Maxim), he could not handle the load, not enough character, too feminine, shook the boy, but you will succeed you are a man!
Thank you!!! I am aware of it....
Does it make sense to use NS if you can do something simpler - just go through history? What is their advantage?
The advantage of NS is the speed of prediction. It takes a long time to train, but it takes milliseconds of time to predict.
Second way, with history and patterns, at every new bar we have to go over the whole history again and check the patterns, it's a long time.
TC keep your voice down about ZZ. Don't say it out loud in this thread, or they will throw tomatoes at you. :-)
Well first, I analyze the data for the period of the window, which I have a floating. This is one.
Secondly. I think it would be possible to analyze my trade history, or rather it would be possible to analyze it, but I don't think it can be applied. I will tell you once again. In order to begin to analyze the market you need a fait accompli. In my case these are signals from the counter-trend strategy. In my case, too. Or you can analyze each bar. The trader has opened a position emotionally, so this event has no regularity of occurrence. How do we know that it is time to analyze the market? When the trader opened the position? Well, then it will be too late as the position has already been opened. So. First, we formalize the moment of decision-making. This is usually the basic strategy.
Last time about ZZ, it is very interesting to analyze the situation that prevailed in the market at its peak - it is strange that NS did not find a pattern ...
Read the article - interesting, but from scratch it is very difficult to begin to realize - it would be interesting to cooperate with anyone...
About the window, if what is in the article, in my opinion, it is not really big, i.e. we expect a movement of much larger size than what is in the window - on less data we plan more - it confuses me.
As for the analysis of deal history, well, that's what I want to understand, maybe this person opens deals by his system - he is a visual person - he takes a decision based on a picture (I mean the impression of information, which may include both news and price patterns), but he is not ready to verbalize it in details. And NS, would be able to put his thoughts on paper, or confirm that there is no logic there.
The first picture shows the pattern 30 bars backwards. It is possible to make 1000 bars, but then even the 10 most similar variants will not be similar to the desired original. And the forecast will still be about 0.
Here's a variation on the 200 bar pattern. As you can see - the scatter of the original is very strong, and it's already hard to call the charts similar.
As for the forecast around zero. A few weeks ago I trained NS and it worked around zero. What was my surprise when I decided to go 0.00050 pips better than the signal. By and large the spread.
After all, if TS is losing within the spread, then why does not the market take away this spread by entering at a price better than the signal at least by the value of the spread. What was my surprise when I saw that equity was smoothly growing upwards. There were slightly fewer trades, which is understandable, but I was very pleased with the balance curve. So. If you have a forecast around zero, it easily becomes profitable when you open a trade at a better price. Even within the spread. IMHO.
Thank you!!! I'm aware of it....
Anyway, if you think about it, don't hesitate to knock right away. Believe me, you will not regret it!
Here is the variation for the 200-bar pattern. As you can see - the scatter of the original is very large, and it is hard to call the charts similar.
Since we already have an indicator, we should use it to create an Expert Advisor. The spread may be large, but the main thing is that the prediction hits the trends. For M1 it will most probably eat away all profits, it would be better to start with something like D1, then H12, H8, H6, H4, etc. and decrease TF while profits will grow.
Last time about ZZ, it is very interesting to analyze the situation that prevailed in the market at its peak - it is strange that NS did not find a pattern...
Read the article - interesting, but from scratch it is very difficult to begin to realize - it would be interesting to cooperate with someone...
About the window, if what is in the article, in my opinion, it is not really big, i.e. we expect a movement of much larger size than what is in the window - on less data we plan more - it confuses me.
As for the analysis of deal history, well, that's what I want to understand, maybe this person opens deals by his system - he is a visual person - he takes a decision based on a picture (I mean the impression of information, which may include both news and price patterns), but he is not ready to verbalize it in details. And NS, would be able to put his thoughts on paper, or confirm that there is no logic there.
Unfortunately, no. NS is not a soothsayer or voodoo. She can't do anything on her own. She only does what is in her, nothing more. The better we prepare data for it, the better it will work.
And the answer is simple. We will know after several bars that there has been a current peak. This is a classic ZZ. But we have one that shows on a zero bar the value of GZ. This one can be used in training and for prediction.
Back in the times of NS the guys in the laboratory made such ZZs. And it kind of worked out even....
Well here they write that the price comes to equilibrium after large deals and news for a few minutes, i.e. something can be caught probably only on M1
I'm lowering the similarity requirement, too, just to find something).
On the subject of prediction, it's a pretty interesting topic. So, we have a paternoster, found in the back of history. Exactly the same as it is now. However, the market reaction to this pattern is not unambiguous. There is up and there is down.
Moreover, as can be seen from the charts, the same pattern has been found several times, so we have several possible outcomes. This is where we need to turn on the Classification. But we will train the network for one pattern. We will label those patterns that caused the market growth as 1, we will label those that caused the market decline as 0. When this pattern appears, we will give the input values at that moment and the NS will tell us if this is a pattern for growth or for fall of the market.
Personally, I see the work in this direction.
The idea itself is very interesting. But if we want to prepare the NS according to this idea, it will be necessary to do it accurately in order not to look into the future etc. When preparing the data for training.
Unfortunately, no. NS is not a soothsayer or voodoo. She can't do anything on her own. It only does what it is designed to do, nothing more. The better we prepare data for it, the better it will work.
And the answer is simple. We will know after several bars that there has been a current peak. This is a classic ZZ. But we have one that shows on a zero bar the value of GZ. This one can be used in training and for prediction.
Back in the times of NS the guys in the laboratory made such ZZs. And it seemed to work even....
I apologize, maybe it's stupid, but with my poor knowledge in this subject I do not see the reason why it is not possible - after reading your article.
I don't understand about the SZ - usual SZ shows the current extremum...
And again RZ is for signal generation, not for its confirmation - i.e. there is always a signal - to sell when the market is rising, but the NS should confirm it, based on the picture of past variations.
I correctly understand that the advantage of NS is that the signal can be confirmed or rejected by a number of patterns, which are collected on the history, must not contradict each other and are checked when a trading signal appears?