Money management strategies. Martingale. - page 11

 
Mathemat >> :

Let me try. The main philosophical question as applied to forex trading sounds like this: "Is there a consistently profitable forex strategy or not?". In relation to it we can offer the following classification of seekers - "highbrow" and others, "simple".

1. The high-browed have heard of the word "martingale" and sometimes even have a pretty good idea of what it is, of Dub's theorem for stopping a martingale (this is just below). They all know that the actual quoting process is very similar to a martingale.

But there is also a scattering of opinions amongst them. Some believe that the market process is 100% proven martingale and you can't make porridge on it. They say, "There is only one strategy that works - arbitrage. Everything else is self-defeating".

And the rest of the highbrow (sly ones) like to be a bit self-deluded and say something like this: "All right, even if it's almost martingale, but no one has strictly proven it. Let's look and see if we find something that works".

2. The simple seekers usually confuse the words "martingale" and "martingale" and do not see the difference. I guess it really is easier to live and hope. Well live the military who believe that the T-72 tank is designed for temperatures down to -700 C, and the cosine phi in military conditions can reach 1.8. On the other hand, the military is practically solving problems that academics cannot even approach, believing that they are impossible even theoretically.

It seems to me that the most profitable position is to be a cunning highbrow. That is, to keep looking for your strategy (knowing that it will be much harder to find than ordinary people think), but knowing at least approximately where it makes no sense to go. But there are traders who make constant profits on the market.

Now the definitions themselves - also very much on the fingers.

3. a martingale is a random process for which prediction in some strict mathematical sense gives a trivial result: the predicted value is equal to the last value of the process. For trading it is absolute death. An example of a martingale is a Wiener process (integral of white noise) or Brownian wandering.

2. Dube (an American mathematician) proved the martingale stopping theorem, saying that the integral of an arbitrary function over a martingale is also a martingale. Translated into trader's terms, it roughly translates as follows: if it were firmly established that a quoting process is a martingale, then any strategy based only on this process has a profit expectation equal to zero when the trading time tends to infinity. Of course, not counting commissions and spreads.

I got it, I understood almost everything you were trying to explain, I even understood that you hinted that I am one of those seekers who confuse the words martingale and martingale (I just did not know the meaning of martingale, so I equated it to martingale because of its consonance). And I think I'm even beginning to understand this martingale. And for simplicity, in order to understand that I understood correctly I will ask a question (but do not laugh): is the backgammon game a kind of process, which is called martingale?

 

I don't play backgammon, so I couldn't answer that, even if I knew how it was played. Well, in order to talk about martingale, you must at least know the value that the process itself takes as a result of a single step.

 
sanyooooook писал(а) >>

I got it, I understood almost everything you were trying to explain, I even understood that you hinted that I am one of those seekers who confuse the words martingale and martingale (I just did not know the meaning of martingale and so I equated it to martingale because of its consonance). And I think I'm even beginning to understand this martingale. And just for the sake of simplicity, in order to understand what I understood correctly I will ask a question (but do not mock): Is backgammon a process, which is called martingale?

Backgammon is not. Though the process depends on chance (roll of the dice), but the result depends on previous rolls and decisions (choices) made by the players. Martingale has no dependence on history. If backgammon wasn't limited by "you can't stand on occupied squares" then winnings wouldn't depend on the player's choices, but would depend only on chance - the roll of the dice and it would be a martingale. And who would play such a game? :)

But in backgammon, it is quite clearly seen that there are better choices and better strategies, leading to a win over a long distance. But it is impossible to prove purely formally either. Such a universal strategy must be formulated. But it is obvious that we shall find another strategy for every one, and there is no universal best strategy - everything depends on the situation and the actions of the opponent. Finding a universal strategy is possible only in the most primitive games.

The same goes for price series and earning in the market. Universal strategy for all markets and working forever is a grail, eternal engine, etc.

Although there are indirect signs that prices are not martingale. But for any this manifestation you can come up with a new martingale model into which these signs fit. Be it gravitation towards certain numbers, heavy tails, autocorrelation of volatility, etc. etc.

The only proof that a series of prices is not martingale is a profitable stack with a significant number of trades so that the probability of luck is minimal. But this is a very intimate proof and the one who has it usually has no incentive to provide it.

In short, martingale is a mathematical abstraction and not a class of real processes.

 
Mathemat >> :

3. a martingale is a random process for which prediction in some strict mathematical sense gives a trivial result: the predicted value is equal to the last value of the process. For trading it is absolute death.

Well, why immediately death? :)

 
paukas писал(а) >>

Well, why die right away? :)

But, after all, Mathemat wrote it in context:

The basic philosophical question as applied to forex trading goes something like this: "Is there a consistently profitable forex trading strategy or not?". In relation to it , the following classification of seekers can be proposed
 
PapaYozh писал(а) >>

But, after all, Mathemat wrote it in context:

The basic philosophical question as applied to forex trading goes something like this: "Is there a consistently profitable forex trading strategy or not?". In relation to it, the following classification of seekers may be proposed

The basic philosophical answer is that there is.

 
paukas писал(а) >>

The basic philosophical answer is to eat.

"To eat or not to eat?", i.e. "To drink or not to drink?", i.e. "To live or not to live?" is the basic philosophical question.

 
PapaYozh писал(а) >>

"To eat or not to eat?", i.e. "To drink or not to drink?", i.e. "To live or not to live?" is the basic philosophical question.

You see, there are no "martingales" in nature. It's all a human concoction.

And in nature almost all processes are inertial, including the movement of cupras.

 
paukas писал(а) >>

You see, there are no "martingales" in nature. It's all a human invention.

And in nature almost all processes are inertial, including the movement of cupras.

You see, I didn't write anything about a martingale. :)

In general, I, as I suppose you do too, consider the martingale to be a theoretical abstraction. There is no martingale in practice.

 
PapaYozh писал(а) >>

You see, I didn't write anything about martingale. :)

This calls for a drink! :)