The market is a controlled dynamic system. - page 378

 
Олег avtomat:

wizard... ;))

More like a pragmatist.

 
aleger:

More of a pragmatist.

That's all right, then.

 

.

 
it's already cold to start the tractors) try starting in the spring instead of at the end of the year
 

.

Yesterday's spikes of 2000 pips -- ~8 times the channel width. Volatility in all its glory.

Corrections made. Let's move on.

 
Олег avtomat:

Yesterday's spikes of 2000 pips -- ~8 times the channel width. Volatility in all its glory.

Corrections made. Moving on.

The problem, apparently, is the same as mine - lack of a desired key "trend/float". It does not take into account the "memory" of the market.

Vaughn, Basishche takes it into account somehow - either through correlation of increments or some other way...

Without the sought key - all efforts are in vain.

 

In short - you need to know exactly whether there are dependent random variables in the quote stream or independent ones. For a correlation coefficient like

is incredibly difficult to determine the sample size N.

I never managed to connect this coefficient to my TS.

I'll have to look for it... And without the key, I repeat, there is nothing to do in the market with ANY strategy.

 

.

 
Олег avtomat:

wizard... ;))

"...Mighty Oleg bowed his head

And thinks: "What's the guessing?
Wizard, you deceitful, mad old man!
I should despise your prediction!..."
 
Олег avtomat:

What's the matter, are you sad or something, Automatista? You're not drinking bitter, are you? Ahem... Is that a good thing?

Okay. Listen up...

I still think there's a rational basis to your system.

However, you don't have the key...

You need to incorporate momentum analysis into your TS, whether you like it or not.

Enter the trade at non-parametric kurtosis <=10 and sit on the fence at >10.

At non-parametric kurtosis <=10 we practically have a Variance Gamma Process type SB, and your TS shows brilliant results at that.

Do you feel it?

Reason: