Not the Grail, just a regular one - Bablokos!!! - page 219

 
Contender:


If you can see that it is possible to build, then why the question "how"?


In my opinion, everything is logical and consistent: if I see that someone manages to build synthetic FI with the right set of parameters, and then trading them to make a profit, then I, in my mind, quite logically have a question "how?" - how exactly does he do it, what is the algorithm, criteria, etc.
 
_new-rena:

the question is an interesting and unresolved one with the aim of getting a "fighting" machine to earn money, so just thoughts:

- need to pick pairs that are close in volatility

the volatility is levelled by lots as i see it

I personally haven't decided for myself yet - where is the starting point (date and time) from which to apply a coefficient to overlap pairs and pass it off at face value?

 
alexx_v:
Who uses what toolkit when choosing FI and coefficients to form necessary synthetics?


I doubt that such toolkit is freely available. As for Instruments, it may be necessary to monitor all available options from the majors(see attachment) and select the most stable ones.

It's probably the same story with directions... overriding all and selecting the narrowest possible corridor.

++++ ++-- +++- ++-+

+-++ +--- +-+- +--+

mirror ones discarded:

---- --++ ---+ --+-

-+-- -+++ -+-+ -++-

Please add if any option from the directions has not been considered.

I have equalized lots by volatility and cost - Either it is wrong or it is not! Maybe it is not the only one to be considered.

Maybe we should not only take it into account.

I analyzed the joker's shares and came to a conclusion that one should not try to reach a cointegrated channel on a long timeframe, several weeks are enough for one trade.

Files:
35cpnvnhb.txt  6 kb
 
No, there's a bloody recycle! I've looked at it many times and discarded it. Perhaps it's time to study it thoroughly...
 

:)

Don't reread your post early in the morning...

 
alexx_v:

just volatility is aligned by lots as far as I can see



you know, you're right.

At least I have some enlightenment now...

Let's say - a lot is some coefficient...

Have you tried putting coefficients on all the pairs so that they are one string stretched in a straight line, and then we'll see? (just born)

you'd probably have to give them the same deviation from a straight line...

after such an operation they will have the same lot and the work boundaries will become clear and proper

 
_new-rena:

you know, you're right.

At least I have some enlightenment now...

Let's say - a lot is some coefficient...

Have you tried putting coefficients on all the pairs so that they are one string stretched in a straight line, and then we'll see? (just born)

most likely you would have to give them the same deviation from the straight line...

After such an operation, they will have the same lot and the boundaries of the work will also become clear and intrinsic



Wren, for God's sake... i even fidget arithmetically once a year, let alone geometrically.... without us...
 
_new-rena:

Have you tried to put coefficients on all pairs, so that they would be stretched in a straight line, and then we'll see? (just born)

This kind of reasoning has no chance of success. Answer this question for yourself: how is your approach better than that of other bidders?

1. Your approach does not rely on information advantage (insider/macroeconomic indicators/equilibrium relationships with other classes of instruments).

2. Your approach does not rely on the speed advantage of access to trading.

3. Your approach does not rely on the structural complexity of the model. A flexible and complex model has many chances to survive in different market regimes, while a simple model will stop working if those who react to changes in the market situation faster than you (you can cheat and take into account the behavior of those who are faster than you, but eventually the system will reach equilibrium and you will stop earning).

 
anonymous:

This kind of reasoning has no chance of success. Answer this question for yourself: how is your approach better than that of other bidders?

1. Your approach does not rely on information advantage (insider/macroeconomic indicators/equilibrium relationships with other classes of instruments).

2. Your approach does not rely on the speed advantage of access to trading.

3. Your approach does not rely on the structural complexity of the model. A flexible and complex model has many chances to survive in different market regimes, while a simple model will stop working if those who react to changes in the market situation faster than you figure it out (you can cheat and account for the behaviour of those faster than you, but eventually the system will reach equilibrium and you will stop earning).



Your approach doesn't rely on anything at all, so it's entitled to life... :-)))

( morgan stanley more often - usually comes true...)

 
zoritch:


Your approach is not based on anything at all, so it is entitled to life... :-)))

Okay :(

( morgan stanley more often - usually comes true...)

As one grandfather said to his prankster granddaughter "better a metre in the hand than a decimetre in the...".