Why is the normal distribution not normal? - page 19

 

getch писал(а) >> Отлично! Ведь рыночное время - мера изменения фин. объема. Не понял, правда, что такое a(n) в ваших рассуждениях?

This is the opening price, which is formed from the minutes. For the analysis I needed to show the dependence of instrument volatility on the well-known parameter - Time Frame. Having a number of minutes, I can generate any TF. For example, I can get TF10 by considering opening prices of one-minute bars in 10-bar steps. Thus, it turns out that the difference between opening prices of bars in TF10 is the difference of bars opening in TF1 - a(i*10)-a((i+1)*10), where i takes values 1, 2...
 

to Neutron


Sergei! I am only now beginning to discover the full scale of your creative personality in you!

..и время исключаю из анали полностью. В нём нет смысла..

and further

And in general, you can't do without time.

:о)))

 
Well, you got it... Bastard!
 

TFN is a time reference, not a price reference.

P.S. I have updated my previous post.

 
Neutron >> :
Well, you got it... Bastard!

>> yeah, but I'll tell you that's a joke, and a good one at that.)

 
getch >> :

About the need to account for human time, I disagree.

So. Our task is to build a TS maximizing only one parameter - profitability (taking into account brokerage commissions). Right?

No! Let me explain why: Suppose we have only two points on the set of experimental points corresponding to the yield parameter - one at maximum, the other - at half yield. The difference between them is that the point at the maximum has one transaction per month on the average, while the second point has one transaction per day.

Question: Which TC parameter is more interesting from a practical point of view?

Answer: The one that brings more income per unit of "human" time. And this is a point with half yield, because the product of 1/2*30 is 15 times greater than 1*1.

It's a slightly exaggerated example (you have to take into account the risks which are less for the point at the maximum) that time does enter into the problem.

 
A conceptual error! Taking human time into account has no effect on the estimation of optimum returns. Again look at the EA.
 

I'm lazy.

I don't want to watch someone else's advice. I'd rather you tell me where I'm wrong. The example I gave is quite illustrative and written in human language, suitable for understanding the essence of the phenomenon.

 
The maximum possible return at any interval will be obtained by traversing ALL local extrema, with a distance between them of at least Spread pips. There is no time limit here.
 

You're smarter than me - I don't understand you.

I'm going to do something else. I'm gonna eat.