Is there a pattern to the chaos? Let's try to find it! Machine learning on the example of a specific sample. - page 30
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
How can 5-10 models determine correctness by themselves? Do you mean average?
Weights can be set depending on the historical accuracy of the training/testing sample.
You can assign weights to combinations.
In general, you can build a tree or another not very complex model and it will determine which signals are the most accurate in combination with which ones.
Or you can take the current history, and if the model is often wrong lately, you can lower its weight - by how much and how - you can fantasise here.
You write "Yes, almost pure values, different depths, different windows, etc.".
There are 2000 signals on the chart, and you have 333 in the description - or I don't understand something again...
Okay, if this is the graph of the last sample, it turns out that the model trained on EURUSD works perfectly on 3 different currency instruments, including cross. I guess it's time for a Nobel Prize!
Thank you for an interesting evening, and all the best to you!
If we take a closer look at the chart, we can see this picture:
A step is the end of a trade and the beginning of a new one. I.e. buy and sell signals are not on every bar, but on the change of direction. Therefore, there were 333 trades on 2000 bars.
On a randome schedule and learning from it?
Yes. The random schedule is formed by 12000 steps. The first 10,000 trains, the rest test. No more sampling.
The weights can be set depending on the historical accuracy on the training/testing sample.
You can assign weights to combinations.
In general, you can build a tree or other not very complex model and it will determine which signals are the most accurate in combination with which ones.
Or you can take the current history, and if the model is often wrong lately, you can lower its weight - by how much and how - you can fantasise here.
You are thinking the right way!!! :) It's a pleasure to read intelligent thoughts ... :)
Bye now for sure.
Regards, RomFil.
If you take a closer look at the graph, you can see a picture like this:
A step is the end of a trade and the beginning of a new one. I.e. buy and sell signals are not on every bar, but on the change of direction. Therefore, there were 333 trades on 2000 bars.
Okay, I get the counting thing. So we identify the trend through the window and follow it.
No, the oscillator is not a collective image, it's real (placed in the basement):
The oscillator itself + points is a forecast. Points appear on the first ticks of a new bar. But the appearance of a point is not a signal to a deal - it is only a warning. Further the further price movement is analysed and only then the decision on the deal is made. By the way, this chart also shows a stop (red mark), which in 98-99% of cases is not broken through - it is a defence against any sharp fluctuations. This is the signal to buy actually ... :)
I have drawn verticals through the price extrema. You can't see the prediction - rather the lag of your oscillator, which is your forecast. Something is wrong here.
And it is impossible to predict random by definition, otherwise it is not random at all.
Are you here to make fun of me?
If not, we're waiting for the signal. I will subscribe if a month will show the same profit graphs as you show here.
I have drawn verticals through the price extremes. You can't see the prediction - it's more like a lag of your oscillator, which is your prediction. Something's not right here.
And it is impossible to predict random by definition, otherwise it is not random at all.
Are you here to make fun of me?
If not, we're waiting for a signal. I'll subscribe if the month will show the same profit charts as you show here.
Good afternoon! I was expecting just such a hit-and-run ... :) I'm not going to explain anything more. It is impossible to predict random, but it is possible to adjust to it. The approach is described above in detail. You can draw verticals by extrema for as long as you like ... Of course, if it gives you pleasure ... I told you everything I wanted to tell you yesterday. Concrete realisation costs money ... :) All the best. And to all non-believers in the grail draw on extremums and believe that the future is predetermined ... :)
The gist of the reply is also expected. :)
Then we are waiting for you on the cover of Forbes or on the Nobel Prize.
Would you sell such a Grail for money?
Did I understand correctly that ML/DL/NN has triumphed in this thread and it predicts (not trading yet) in plus with zero spread/commission but on randomly generated inputs?
and this is being seriously discussed...
I understood correctly in this thread ML/DL/NN has triumphed and it forecasts (not trading yet) in plus with zero spread/commission but on randomly generated inputs?
and this is being seriously discussed...
No, he's not. He's just here for the fun of it. Maybe a grail to sell for cheap.
There's no celebration. It's the usual 50/50. Maybe with a slight edge in our favour.)