Machine learning in trading: theory, models, practice and algo-trading - page 1725

[mytarmailS]

Maxim Dmitrievsky:

I'm sure you did it just yesterday, after my posts and articles... maybe a little earlier

Ahahahaha)) You fucking narcissist))

Here's a folder with the script and everything about this topic

[Mihail Marchukajtes]

Rorschach:

Now I'll tell you something even cooler, you can make money with no effort at all))

So where's your shaitan machine a la printing press?

[[Deleted]]

mytarmailS:

You're a fucking narcissist.)

Here's the folder with that script and everything about it

ok))

so did you make any progress? or did you drop it?

[mytarmailS]

Maxim Dmitrievsky:

ok )))

Did you develop the theme? Or gave it up?

I abandoned it, although the topic seems to be working, I want to predict everything)))

[Rorschach]

Here you go))

To make a long story short,

any system is taken,

collects statistics on the length of losing and winning streaks,

and then calculate the mania management.

Tried to get into the math, it seems there is a mistake in the calculations, but it worked for the author.

[[Deleted]]

mytarmailS:

I gave up, although the topic seems to work, I want to predict everything)))

If I start a reader by trying all the hours and minutes by some criteria, and then combine all the best ones in the pile. I wonder if such a mess would work or not, for signals are very short, i.e. minute ones.

ran the calculator on a sample of 10 years, let it calculate everything

Hour:  0  Minute:  2  Score:  0.9090314522857109
Hour:  0  Minute:  3  Score:  0.9108283726238224
Hour:  0  Minute:  4  Score:  0.9108501024096574
Hour:  0  Minute:  5  Score:  0.9026031634097038
Hour:  0  Minute:  29  Score:  0.9390718447033435
Hour:  0  Minute:  30  Score:  0.9301369431625292
Hour:  0  Minute:  31  Score:  0.9394297165567186
Hour:  0  Minute:  32  Score:  0.9148751198429192

[Mihail Marchukajtes]

Well, let's put it this way, first of all, money management can pull not every strategy. If the strategy is fiercely unprofitable, it will not help you.

The main point of money management is to improve the performance of the system, that is, to straighten the balance curve at an angle of 45 degrees. Here I have MM improves such an indicator as profitability, which is also important in real trading ....

[mytarmailS]

Maxim Dmitrievsky:

And if you run a counting machine by trying all the hours and minutes by some criterion, and then combine all the best ones into a pile. I wonder if such a mess would work or not, because the signals are very short, minute-long.

Nope, only those three hours work (and it wasn't always that way, it appeared relatively recently), the more you start shuffling, the more everything "floats" into uncertainty ...

Since these "working hours" are in fact the opening of a new daily candle and the closing of the old one, I think the pattern is in the opening of the positions of those who trade on the diaries, but why is it always in one direction? Or maybe the banks just do something systematic

Do you know what you can do, you can try to filter by months and days of the week

[Rorschach]

Mihail Marchukajtes:

Well, let's put it this way, first of all, money management can pull not every strategy. If the strategy is fiercely unprofitable, it will not help you.

The main point of money management is to improve system indicators, i.e. to straighten the balance curve at a 45 degree angle. I have MM improving such indicator as profitability, which is also important in real trading....

I took a system dangling at zero. It turns out that equity serves as an oscillator. When it deviates to a certain level, it enters with real money.

[[Deleted]]

The question is whether it is possible to trade every minute separately ) whether there is a profit more than the spread