And recognition is no longer about recognising a trend-flat change, but about predicting it.
In my opinion, the idea of being able to predict such a shift is a dangerous illusion. It is possible to get a few right predictions in a row, but it will not help to build a system that works for a long time.
It is only possible to react to such a shift, and it will always be a matter of adjusting the balance between first- and second-order error probabilities (shift misprediction and shift omission).
darirunu1: I am sorry, it is none of my business, but why are you looking for pictures but not for what makes these pictures or what is in it. Knowing that in essence the images will not matter.
What makes these 'images' is hidden from the trader. That is why the images are searched for, so that the "Creator" can be deduced from the "Creation". But I will not continue this topic - it is too voluminous. I'm afraid that I too will be labeled as a schizophrenic).
In my opinion, the idea that it is possible to predict such a shift is a dangerous illusion. It may well be possible to make several correct shift predictions in a row, but it will not be possible to build a system that works long enough.
You can only react to such a shift, and it will always be a matter of adjusting the balance between first- and second-order error probabilities (shift misprediction and shift omission error).
It is clear that this is a probabilistic process. A trader needs to be familiar with probability theory.
Imagine that a person's life is one graph, and since people are different, the graphs are different, just like currency pairs. What do all people have in common? Well, the first is that they are born and die, that is, a trend begins and ends. Then we do not know, but everyone has childhood, youth, maturity and old age.
Now let's imagine over the years which ages are covered by childhood, adolescence, maturity and old age, so we will find a common denominator for all people.
candle shapes and combinations are not about that at all. This is also a sandbox issue.
Nevertheless, we will always have to deal with some specific decision rule on the feature space we use. One way or another, it will come down to a function of historical prices (and possibly news). Thus, the use of some kind of indicators from price is unavoidable.
What makes these 'images' is hidden from the trader. That is why the images are searched for, so that the "Creator" can be deduced from the "Creation". But I will not continue this topic - it is too voluminous. I am afraid that I may also be labeled as a schizophrenic).
By and large, any algo is a calculation of the "Creator" by his "Creation".
By and large, any algo is a calculation of the "Creator" by his "Creation".
In my opinion, it is the AI that sits in people's brains that prevents us from solving the problem. This is like a crowbar with no resistance if there is no other crowbar.)
It's like you remember the adventures of Electronica. It was like, "Eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee, don't need Eeeeeeee."
Allow the use of cookies to log in to the MQL5.com website.
Please enable the necessary setting in your browser, otherwise you will not be able to log in.
You are signed in. Two-factor authentication has been disabled. Set it up again and scan a new QR code, or continue with two-factor authentication disabled.
And recognition is no longer about recognising a trend-flat change, but about predicting it.
In my opinion, the idea of being able to predict such a shift is a dangerous illusion. It is possible to get a few right predictions in a row, but it will not help to build a system that works for a long time.
It is only possible to react to such a shift, and it will always be a matter of adjusting the balance between first- and second-order error probabilities (shift misprediction and shift omission).
I am sorry, it is none of my business, but why are you looking for pictures but not for what makes these pictures or what is in it. Knowing that in essence the images will not matter.
What makes these 'images' is hidden from the trader. That is why the images are searched for, so that the "Creator" can be deduced from the "Creation". But I will not continue this topic - it is too voluminous. I'm afraid that I too will be labeled as a schizophrenic).
In my opinion, the idea that it is possible to predict such a shift is a dangerous illusion. It may well be possible to make several correct shift predictions in a row, but it will not be possible to build a system that works long enough.
You can only react to such a shift, and it will always be a matter of adjusting the balance between first- and second-order error probabilities (shift misprediction and shift omission error).
It is clear that this is a probabilistic process.
A trader needs to be familiar with probability theory.
I understand. So the image is shapes and forms of candles and their combinations?
candle shapes and combinations are not about that at all. It's also from the sandbox category.
candle shapes and combinations are not about that at all. This is also a sandbox issue.
Nevertheless, we will always have to deal with some specific decision rule on the feature space we use. One way or another, it will come down to a function of historical prices (and possibly news). Thus, the use of some kind of indicators from price is unavoidable.
What makes these 'images' is hidden from the trader. That is why the images are searched for, so that the "Creator" can be deduced from the "Creation". But I will not continue this topic - it is too voluminous. I am afraid that I may also be labeled as a schizophrenic).
By and large, any algo is a calculation of the "Creator" by his "Creation".
By and large, any algo is a calculation of the "Creator" by his "Creation".
In my opinion, it is the AI that sits in people's brains that prevents us from solving the problem. This is like a crowbar with no resistance if there is no other crowbar.)
It's like you remember the adventures of Electronica. It was like, "Eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee, don't need Eeeeeeee."