On the unequal probability of a price move up or down - page 160

 
Aleksey Nikolayev:

What if, in addition to the non-stationarity that cointegration can help with, there may be other types of non-stationarity? This is, for example, what in econometrics is called structural jumps.

Many things can happen in the world against which one has no protection.)

 
Maxim Kuznetsov:

Maybe instead of guessing on a triangle, we could use real statistics?

For example, for the pound to take the history of quotes, a report like the attached: somehow convert the turnover from USD to GBP, then on the hypothesis that the total number of pounds per month does not change much count the weighting coefficients and identify the overbought/oversold.

I suppose it will come down to adding regression on these turns to the autoregressive equation. The number of model coefficients will increase and, as usual, we will have to investigate whether the model has really improved or whether a trivial retraining has taken place.

 
Grigori.S.B:

Co-integration in the forex market is far from reality.

But it is possible to construct synthetic instruments with any predetermined properties. Of course, there will still be the problem of stability. But that's the way the world works. No, there was not and never will be stability in this world).

 
khorosh:

Many things can happen in the world against which one has no protection.)

The most important thing in our business is to get away in time to close a position, not to sit on losses, as some magicians do).

 
Aleksey Nikolayev:

The most important thing in our business is to be able to close a position on time, not to sit on losses like some magicians).

Quite right, we should not disregard this opportunity for the illusion of making a profit whatever the cost. By losing less, we save more).

 

The robot has just worked out )


 
khorosh:

But it is possible to construct synthetic tools with any, predetermined properties.

Really? Even with the property of stationarity?

 
Evgeniy Chumakov:

The robot has just worked out )


Yay!


 
Grigori.S.B:

Really? Even with the property of stationarity?

khorosh:

But it is possible to design synthetic tools with any predetermined properties. Of course, there would still be the problem of stability. But that's the way the world works. No, there was not and never will be stability in this world).

Have you found any particularities in the interpretation of the terms: stationarity, stability, sustainability?
 
Grigori.S.B:

Really? Even with the property of stationarity?

Ooh, okay.

+