From theory to practice - page 148

New comment
 
Alexander_K2:
Once again, for the particularly gifted, I say - the algorithm is not mine, it's Vizard_'s.

You must have misunderstood his algorithm. There was something about some statistical calculations, and when they are out of tolerance it means that the trend is about to reverse.

"Some statistical calculation" - either he didn't write about that, or I didn't notice it, but I doubt it's a trivial moving average. Try putting a different formula in there.

 
bas:

It doesn't matter, you don't realise your own algorithms either, for example you tried to give WMA weights by tick size, although such MA is 99% the same as SMA, and this is obvious to anyone who knows how to analyse data and think logically. I'm not being mean, if anything, just suggesting how not to waste unnecessary effort) Even though you don't hear others)

No, it's important. Sorcerer (Vizard_) has never written the utterest rubbish as you have. It remains to really get his point across, rather than amusing the honest folk with the idiotic Mash-ups you've stuck to.
 
Dr. Trader:

You must have misunderstood his algorithm. There was something about some statistical calculations, and when they are out of tolerance it means that the trend is about to reverse.

"Some statistical calculation" - either he didn't write about that, or I didn't notice it, but I doubt it's a trivial moving average. Try putting a different formula in there.

I think he went beyond the trivial stuff like rolling back to the average. I think he slips all the increments into the NS input and fiddles with unique things there. And here he's just having fun.
 
Wizard2018:

Why do you need formulas on the second sheet? There are copied columns A,N,M,O from another sheet.Column A is the Bid price, for N, M, O the formulas are there.


5. Move to the Sheet 2 of the table.

I have copied columns A, N, M, O from the AUDCAD tab to it, starting from the row 15625


Here is the algorithm point by point.

https://www.mql5.com/ru/forum/221552/page135#comment_6366221

Thanks, I think I understand a little bit, which makes the absurdity of the idea stronger. Also the process of reducing the process to Markovianity seemed to me to be essentially different than the primitive one shown as "reducing to Markovianity" in Excel.
 
Wizard2018:
Was, download the file and read its description just above. All the formulas are in the cells. What else do you need? Algorithm from example Excel file and description is completely clear. Ihave already copied formulas to my self-written terminal and all calculations and "pictures" have certainly coincided. I have faced the lack of iron resources for calculations of large tick volumes.


Forum on trading, automated trading systems and trading strategy testing

Difference calculus, examples.

Vizard_, 2018.01.19 07:33

Polynomials, splines, Gaussian processes...
Blue dots are training dots, red dots are test dots. Generate a bunch of any curvature on the blue ones, test
on a metric you like on the red ones and select the best one. You can randomly remove some of the blues...


Does your self-written terminal do these calculations?
 
Aleksey Panfilov:


Is this the kind of construction your self-written terminal does?
I read somewhere that Warlock has several Grails. I think he shares them in each of the threads. The truth is vague and there's probably a good reason for that. And all we have to do is understand what he says, well, and use it of course.
 
Alexander_K2:
I read somewhere that Warlock has several Grails. I think in each of the threads he shares them. The truth is vague and there's probably a good reason for that. And all we have to do is understand what he says, well, and use it of course.

Mm-hmm.

So he's teasing. ))))

"Wings? Legs? ... Tail!!!"

 
revers45:

It's like a final session in Vasyuki and an adequate grandmaster has already realised it's time to run away...

Well in the beginning of the session was declared quite a non-trivial idea - to simulate the evolution of statistical characteristics by Fokker-plank. But it seems that in view of the complexity of the problem, Grandmaster gradually moved to a simple mach.
 
sibirqk:
Well in the beginning of the session it was declared quite nontrivial idea - to model evolution of statistical characteristics by Fokker-Planck. But, it seems, in view of the complexity of the problem, the grandmaster gradually moved on to a simple mach.

Nothing of the sort - I haven't given up on my idea. We are dealing with a wave price function (price probability density function), which we analyse in time and whose movement is described by the Fokker-Planck equation. This can be considered proved and we will not dwell on it once again.

But further on, in view of newly discovered circumstances, things get so non-trivial, that I am now simply following the advice of one of my teachers, who said: "If it gets complicated - read Feynman". I'm sitting and reading.

Starting to understand now?

 

I think the original formulation of the problem, the movement of the market as an aggregate movement of quantum particles, is wrong.

IMHO it is necessary to model the aggregate state of cellular automata experiencing emotions regarding their personal equity.

Equity is on the plus side - greed.

Equity goes up - buy and hold

Equity falls to 50% of max profit taking.

Equity in minus - fear.

Equity goes up - close with no loss.

Equity goes down - close on SL


This is where formulas for market movements should be built from.

If the market goes up, you need to buy, if the market goes down, you need to sell (if you do not have a position in the cage).

1...141142143144145146147148149150151152153154155...1981
New comment
Reason: