Machine learning in trading: theory, models, practice and algo-trading - page 777

 
I want, I want... MMMM recruit investors for a million dollars....... :-)
 
I decided to bomb the masses and as if by chance the signal is unprofitable :-( and quite strongly. I'm always this way, just when I'm about to get rich I'm just about to get rich :-(. In the hope that in Asia we'll bounce back at least a little bit...
 
Does anyone happen to know what kind of bastard is pulling the eu and its like down???? So... just asking...
 
Mihail Marchukajtes:
Does anyone happen to know what kind of bastard is pulling the eu and its like down???? So... just asking...

someone's risky, or rather the net

No turkey, NS, etc. will ever be able to show this and unravel/train/prepare/etc.

 
Mihail Marchukajtes:
Does anyone happen to know what kind of bastard is pulling the eu and its like down???? So... I was just asking...

Forgive me, for Christ's sake, I am a sinful fool))) I didn't just say "Icarus")))

 
Renat Akhtyamov:

Somebody's risk is off the charts.

i.e., i don't know how to make it so i don't know what to do and what to do with it.

I call it a mistake when the TS gets a minus. As a rule in a normal TS minus should not be large, but here ... BUCKY :-(

 
Mihail Marchukajtes:

Yes... counts the standard deviation of the comulative delta

Are you sure? Judging by the code, you count the MA of the cumulative delta (hundreds ... thousands of units) and find the difference with the price (1.41 ... 1.42).
dAmount+=(dAPrice-dMovingAverage)*(dAPrice-dMovingAverage); where dMovingAverage is the MA of the cumulative delta

Price has negligible effect on the result. I think the MA of the cumulative delta is enough. Or just the cumulative delta.

 
Mihail Marchukajtes:

I call it a mistake when the TC gets a minus. As a rule in a normal TC minus should not be large, but here ... BUCKY :-(

I only wish it were so bad

they just become many, i.e. more

 
Renat Akhtyamov:

Somebody's risk is too high.

no turkey, NS, etc. will ever be able to show it

My TS worked like this:


 
elibrarius:
Are you sure?... Judging by the code, you count the MA of the cumulative delta (hundreds ... thousands of units) and find the difference with the price (1.41 ... 1.42).
dAmount+=(dAPrice-dMovingAverage)*(dAPrice-dMovingAverage); where dMovingAverage is the MA of the cumulative delta.

The price has negligible effect on the result. I think the MA cumulative delta is sufficient. Or just a cumulative delta.

Or then subtract the MA from the cumulative delta from the cumulative delta itself.
Then the deviation will be

Reason: