Looking for patterns - page 50

 
Vitaliy Maznev:

But at the expense of it, this future can be limited to the point where there are no options left other than those given by the system. That is, I can define it as "there is no future". In the most fatal sense of the expression.

I can see that the ring is already narrowing down and quite fast. I may not see the slamming down any more. But it's the best case scenario:)

 
How come it doesn't work when your picture has 6 wins and one loss, put another picture where it doesn't work
 
Uladzimir Izerski:

I can see that the ring is already narrowing and quite rapidly. I probably won't see the slamming already. But that's the best case scenario.)

No one will see the slamming. Because it has already happened. And from here I deduce the perspective of AI and the lack of it in society.

 
Yury Stukalov:
Put another picture where it doesn't work

It's easier to make an EA and run it on history. And you will see by the large number of trades.

 
Aleksei Stepanenko:

It's easier to make an EA and run it on history. And you will be able to see by the large number of trades.

Do it, and first we'll see if the algorithm is correct, and then the result.

 
Aleksei Stepanenko:

Sorry, didn't answer your question but remembered it.

Let us create an array where each element of the array is an extremum. The next element is the opposite extremum. We collect all information for each extremum: price, date, distance to the previous extremum, time period to the previous extremum. As a result, we have a set of accurate data from which we can calculate the amplitude and period of oscillations. Maximum, minimum, average, median, and get probability plots. And all this is accurate, one-to-one with the price. And what to use to register an extremum, there are options here: minimum distance, minimum interval, more can be invented.

So this is for history analysis. And if we need to identify the last extremum online. You will identify it with a lag as well. Because you imagined that the instrument M lags but the extrema do not lag.

 
Aleksei Stepanenko:
What about the flat, when it goes back and forth? And when the pullback was there, but it did not touch the average? On some slice you will get lucky, but statistically it doesn't work. Such is the case.

Flat is a casino with small stakes. You're not going to lose everything. Only on spreads.

In the casino, you can bet on UGAD.

In the markets you can bet at random.

But it is on hope, on a random result.

But this variant is not suitable for working on Fr.

To understand where the price will go, we should first distinguish the trend from the wave, and it is even more important to analyze the FA.

 
khorosh:

Ma is lagging, but extrema are not.

Yes it does, it's for history. So it is important to have accurate data on history in order to draw conclusions. And the moving average is of little use here. And no one knows the future, there is no such tool and will never be, there is no argument here.

 
Yury Stukalov:

do it, first let's see if the algorithm is correct, and then the result

State the exact conditions.

Or maybe you can write it yourself? I don't have much time.

 
Aleksei Stepanenko:

Yes it does, it's for history. So it's important to have accurate data on history to draw conclusions. And sliding is of little use here. And no one knows the future, there is no such tool and will never be, there is no argument here.

You're right that no one knows the future. There's just one small thing. Inertia. It's never been eliminated. And it's a big deal.