Pair trading and multicurrency arbitrage. The showdown. - page 70

 
It is necessary to fix the divergence, then put buy stops up on the pair that is lower and sell stops on the one that is higher - then the first thing: either the pairs will converge, or they will not converge and go in one direction as they are correlated. Both will be plus
 
Vladislav Vidiukov #:
It is necessary to fix the divergence, then put buy stops up on the pair that is lower and sell stops on the one that is higher - then the first thing: either the pairs will converge, or they will not converge and go in one direction as they are correlated. Both will be plus.

That's what I call an inverse pairing. Haven't done it yet. It's a long way off. And there's a lot of options for catching the original bifurcation...

 
Roman Poshtar #:

That's what I call a reverse pairing. Haven't done any specifics yet. It's a long way off. And there's a lot of options for catching the original sliding....

I don't think you understand. It is necessary this direct pairing, taking into account the fact that if the pairs do not converge, they will go in one direction and there will be a profit of at least 1. There you need to put a grid of buy and sell stops.
 
Vladislav Vidiukov #:
I don't think you understand. It needs to be a direct pairing, taking into account that if the pairs do not converge, they will go in one direction and there will be a profit of at least 1. There you need to put a grid of buy and sell stops.

I'll figure it out, don't worry.)

 

Sculpting by hand is not bad. But you still need to write a robot to test it.


 
Roman Poshtar #:

Sculpting by hand is not bad. But you still need to write a robot to test it.


go ahead - write the conditions for input-output...

with variants...

 
Roman Shiredchenko #:

come on - write up the conditions on the input output...

with options...

Ask the author of the indicator above. If he allows to post it. But so he has described everything.

 
Vladislav Vidiukov buy order on the 2nd pair and wait for the rise on the 2nd pair. It is not necessary to wait in ABSOLUTE numbers of convergence! The pairs go PRIBLY synchronously, not absolutely. Absolutely is cointegration, which does not exist in Forex!

In fact, all movements on all pairs (on currencies, scientifically pairs are vapour combinations) are strictly synchronous. Simply because there is no triangular arbitrage A->B->C->(A+profit).

If profit is maximised, then any pair/unpair/goat trading in the currency market boils down to trading one currency against another. Roughly - the right choice is leader vs. loser.
When the moment of "beginning of the sliding" and continuation of leadership/looser is caught - this is trend trading. When you catch the end of the process, i.e. the beginning of "convergence" - it is counter-trend.

If the risk is minimised, it is basket trading and quick shifting of hot/cold, which again is leader vs. loser.

 

musings...

I think that if we translate the logic of this single-character owl
(it is remarkable because it has no parameters at all, and is not based on AI/MO
counts statistics and trades on volatility+periodicity).

If you trade "in a bundle" and in a similar chart, you will get good results.
you will have to turn your brain :-)

 
Roman Poshtar #:

I'll figure it out, don't worry.)

You don't understand. It is necessary to wait for sliding and put buy stops at the lower pair and sell stops at the upper pair, then 1) pairs converge - profit. 2) pairs correlating go in the same direction - profit. 3) pairs diverge even more, but this will not happen as they are correlated and will go in one direction - profit. That is, in 90% of cases there will be profit
Reason: