Machine learning in trading: theory, models, practice and algo-trading - page 3222
![MQL5 - Language of trade strategies built-in the MetaTrader 5 client terminal](https://c.mql5.com/i/registerlandings/logo-2.png)
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
Anyway, it didn't work out here.
I have published the scripts, so you don't need to invent anything yourself to try new data.
Well, then your graph isn't random.
Yeah. There's half a blog post on the subject.
The screen shows that there are 10-20 times more transactions (almost 20K). Some generation settings are obviously grail. Apparently, you make Avg-price fluctuate a lot.
This is most likely the case, otherwise there would be no need to add a six. It didn't go anywhere in its generations.
There are two generations in the screenshot. Random curveballs. They're useless so far.
It probably is, otherwise there would be no need to add a six. It didn't go anywhere in its generations.
There are two generations on the screen. Random curveballs. They're useless so far.
I also noticed that after the generation, the spread is much larger.
I'll check later, maybe it's a mistake.
The only thing is, we've got to do something to avoid that.
Here is this case.
On the left there are relatively few vcrops of the original, on the right there are more.
On the left.
Right.
You can't tell that visually from the graphs. Roughly speaking, the only reason I understand it is because I understand the principle of the generator.
There are relatively few vcrops of the original on the left, and more on the right.
I tried to take the best set on such a pseudo-generation and run it on the original. Got a straight line with a bigger profit.
In general, it's not clear.
it generates some rubbish compared to the original, there are no errors in the code, I think
That's why the spread is bigger, you have to multiply the generated value by 0.3. I don't know why this is so :) Maybe because of rare emissions of the original, they shift the weights
CSV: time bid ask.
It seems to me - milliseconds are not restored correctly. I compared CSV with tick history in the terminal.
I don't get it.