Thoughts on the random - page 5

 
prikolnyjkent: Everyone is well aware that the nature of the statistics will simply change over time.

Then what is the point of researching history if the nature of the real will change anyway?

Prikolnyjkent:And it is most logical to use as many independent processes running in parallel as possible.
History, forward and real do not run parallel and independent. They run sequentially and independently. What parallel independent processes are you talking about?

Prikolnyjkent : The main thing is to "overthrow from the pedestal" the idea that trading is possible only due to the ability to predict the direction of quotes.
I do not understand, what is wrong with being able to predict the direction of quotes?
 
alexeymosc:

Sorry, colleague, but that literally means admitting the unpredictability of the process, its randomness, hence. And that only leads to a flop sooner or later.


And here, colleague, I beg to differ with you...

Don't discount a component that is quite rich in earning potential, such as Vibrations (!). There is MORE "life energy" in the fluctuating statistics of your trades around the 50-odd spot than the spread can eat up. And "don't sweat" the mystical "long unbroken series of failures". It's only inevitable at infinity. And you... How many trades in one "process" per year do you get? (you don't have to answer ;-)...

 
LeoV:

Then what is the point of researching history if the nature of the real will change anyway?

Well, personally, I used history to estimate the starting parameters of the system.

History, forward and real do not run in parallel and independently. They run sequentially and independently. What parallel independent processes are you talking about?

I am talking about parallel processes in real trading in several "directions" (different TS, different instruments) ONCE.

I do not understand, what is wrong with being able to predict the direction of quotes?

It is not bad at all. Simply, it is not the only source of profit. That is all...

 
If you learn to predict fluctuations, why not.
 
alexeymosc:
If you learn to predict fluctuations, why not.


I will only be happy for you...

But as long as you learn to predict, a dumb mathematical mechanism can pull profits out of fluctuations now... albeit smaller ones...

 
prikolnyjkent:
... Don't work with quotes statistics, but with statistics of EXPECTED deals (opened on the signals of some TS), ... and do not look for direction, but for the VOLUME of a new position ... (no thanks ;-)


)) hmmm a la PAMM investor)) ...) You can also work with a system's trade statistics, created on a "game" with trade statistics of another system, etc... You will get a kind of investor of a certain level.

Just a question - why do you think it's easier to play on equity... of, say, your system is easier to play than kotir? How is the statistics of the outcomes "more meaningful" in the context of profit prospects than the same kotir? It's just one of the filters.

 
alexeymosc:
I've already checked it, but using a different method.


What method?

 
nesssesary:


)) hmm a la pamm investor)) ... You can work with a system's trade statistics, created on a "game" with the trade statistics of another system, etc. You will get a kind of investor of a certain level.

Just a question - why do you think it's easier to play on equity... of, say, your system is easier to play than kotir? How is the outcome statistics "more meaningful" in the context of profit prospects than the same kotir? It's just one of the filters.


Outcome statistics have one very attractive feature - they fluctuate around one... and strictly defined level (49% success rate), which cannot be said at all about quotes...
 
prikolnyjkent:

Outcome statistics have one very attractive property - they fluctuate around one... and strictly defined level (49% success rate), which cannot be said at all about quotes...

Can you say that about an ordinary oscillator?
 
alexeymosc:
And I'm happy for you. The Kalashnikov rifle is also simple and reliable, except that its reliability has been tested on millions of weapons over decades. Actually, that's not what I wanted to say. My mind is blurred, I can't put it together. I want to find probabilistic proof that the market isn't random. I've already tested it, but by a different method.


Prove that it is random.