From theory to practice - page 1665

New comment
 
Renat Akhtyamov:

alignment of the number of ticks per period

To the point.

 
https://oxfordstrat.com/data/volatility-clustering-2/
Volatility Clustering: Alternative Methods of Filtering.
Volatility Clustering: Alternative Methods of Filtering.
  • oxfordstrat.com
Concept: Large price moves tend to be followed by large price moves, and small price moves tend to be followed by small price moves (Volatility Clustering). Research Question: Can we improve performance of the original volatility clustering model via standard deviation filtering of large price moves? Specification: Table 1. Results: Figure 1-4...
 

Martin Cheguevara:

.... more than you give it won't fit.

and you can't give more than you can handle, there are bottomless pockets ;))))

so you don't have to build it under the ...

there's a word for it.

well.... and forgotten, I guess.

;)

 
Maxim Dmitrievsky:
h ttps://oxfordstrat.com/data/volatility-clustering-2/

Serious people - they refer to Occam and Popper) the founder studied at the Moscow Power Engineering Institute).

It is necessary to buy strategies from them urgently, before they sell out all of them) I will have to switch to matlab, though

 
Renat Akhtyamov:

analyse (from my message to Jura), really without explanation:

Oh, hell yes of course. 1-2 pt by four digits of profit.......... And if you also take its ratio to the drawdown imho is a pity.

This is when the execution is good; if the amount increases, then one more rake will appear. The strategy has no money capacity.

I saw what I saw, maybe I missed something))))

 
Renat Akhtyamov:

either equalising the number of ticks per period
or the number of readings

To be more exact, with this quote reading method we obtain practically a stationary process. I.e., for any time period, we have the amount of events satisfying the Poisson distribution. The dispersion channel is aligned relatively to the average and does not depend on bursts of activity on the market. Another matter is that giant price spikes are still present and a fully stationary Poisson's process still cannot be achieved... The question of applying an additional decay indicator remains and I cannot yet think of a better one than the kurtosis of the incremental distribution.

PS Somewhere CheGevara had a picture with the distribution of increments, where trend and flat zones are marked. I can't find it... But, I totally agree with her.
 
Alexander_K:

More precisely, with this method of reading quotes we get a practically stationary process. That is, for any time period we have the number of events satisfying the Poisson distribution. The dispersion channel is aligned relatively to the average and does not depend on bursts of activity on the market. Another matter is that giant price spikes are still present and a fully stationary Poisson's process still cannot be achieved... The question of applying an additional decay indicator remains and I cannot yet think of a better one than the kurtosis of the incremental distribution.

PS Somewhere CheGevara had a picture of distribution of increments, where trend and flat zones are marked. Can't find it... But, I totally agree with it.

I didn't think he'd do it either.

 
vladevgeniy:

Yeah, I don't know. 1-2 pts on a four digit profit.......... And if you take its ratio to the drawdown, it's a pity.

This is when the execution is good, and if the amount grows, then one more rake will appear. The strategy has no money capacity.

I saw what I saw, maybe I missed something))))

Look at the picture again.

The point is that legal entities provide transactions to individuals.

and physical persons are left with dead ends and legal entities, look it up.

i'm trying to do the same thing but tester blows out based on stopout - and i don't have one

I'm trying to do the same thing but the tester rolls up

I try the opposite (on the physical side, all is normal)

try again on the yur side of the game - bang and stop...

;)))

 
Renat Akhtyamov:

the point is that legal entities secure transactions for individuals

and physical entities are left hanging and legal entities are left hanging.

I'm trying to make a similar one, but the tester blows out on a stopout, and there is no stopout

I'm trying to do the same thing but the tester rolls out

trying the opposite (on the physical side, all normal)

I try to play on the legal side again - bang and stop...

;)))

Legal entities are funds, they hang for a long time. Individuals are gambling, hence the spread on OI. Nobody's providing anything to anybody, calm down already.

neither the former nor the latter have any effect on the sentiment of futures in general, because futures have no effect on spot
 
Maxim Dmitrievsky:

Legal entities are foundations, they hang on long term. Individuals are gambling, hence the spread on OM. No one is providing anything to anyone, just calm down already

Look carefully, look where the price was going, change the theory and look at what is happening in practice, but with your own eyes, not the eyes of others.
1...165816591660166116621663166416651666166716681669167016711672...1981
New comment
Reason: