Machine learning in trading: theory, models, practice and algo-trading - page 3072

 
Maxim Dmitrievsky #:

I've kept the signs under wraps for a while, cos I've had a long time to choose.

Do they all have a solid foundation? In the sense that the reason why they influence the exchange rate of a given currency is known?

 
Evgeni Gavrilovi #:

Do they all have a solid foundation? In the sense that there is a known reason why they affect the exchange rate of a given currency?

Unfortunately, no. Derivatives of prices, chosen empirically. No causal meaning is invested.

 
Valeriy Yastremskiy #:
Majors. Except eurodollar. To choose from.

NZDUSD hours, 2000 to 2010 OOS.

You can put stops and take-outs at the opening prices to make it faster.

I'm throwing it here, because I think I'll be crazy to give it to everyone in turn, if they ask for it :)

There is some failure at the junction of training and OOS, but otherwise it looks good. I'll try to do better later, I'm improving my approach.

Files:
 
Maxim Dmitrievsky #:

NZDUSD hours, 2000 to 2010 OOS

You can lock in stops and takeouts at open prices to make it faster.

I'm throwing it here because I think I'd be crazy to give it to everyone in turn if they ask me to :)

There is some failure at the junction of training and OOS, but otherwise it looks good. I'll try to do better later, I'm improving my approach.

Thank you. As I read more and more I come to the idea that pure increments on one instrument are not quite the right thing. And maybe even the analysis of all instruments correlated with it. Although maybe something can be picked up. In general, the goal as in shares, whether the rate is fair or not, is more sympathetic to me at the moment. But I have a problem with the signs.

Thorpe described such a system, allocated weekly or 2-week trends, took 10 or 20 per cent of the upper on sell and lower on buy. And it worked) True, at Thorpe they took more than 1000 instruments for analysis, at Renaissance more than 6000. )))) And then began to improve dividing by industrial categories and external factors relative to these categories.

 

Oh that forex ))) and its fictitious overnight spreads


 
fxsaber #:

Do you have overnight spreads?

 
Maxim Dmitrievsky #:

NZDUSD hours, 2000 to 2010 OOS

You can lock in stops and takeouts at open prices to make it faster.

I'm throwing it here because I think I'd be crazy to give it to everyone in turn, if they ask :)

There is some failure at the junction of training and OOS, but otherwise it looks good. I'll try to do better later, I'm improving my approach.

In the __MACOSX folder ? is that for apple ? or should I move the folder too ?

I didn't move it. It works) after 22 years is not very good. Optit set) sl 100 - 20000 in 100, tp 100 - 5000 in 50).

 

Who has tried this approach like the one in the video?


 
Valeriy Yastremskiy #:

In the __MACOSX folder ? is that for apple ? or is the folder to be moved too ?

I didn't move it. It works) after 22 years is not very good. Optit set) sl 100 - 20000 in 100, tp 100 - 5000 in 50).

and I need to tweak it, I've hardly ever trained on this pair. Later I will try.

 
Maxim Dmitrievsky #:

and I need to tweak it, I've hardly ever trained on this pair. I'll try it later.

Since 19 till today the best profit is -39 quid).