Obtaining a stationary BP from a price BP - page 24
You are missing trading opportunities:
- Free trading apps
- Over 8,000 signals for copying
- Economic news for exploring financial markets
Registration
Log in
You agree to website policy and terms of use
If you do not have an account, please register
grasn писал(а) >>
And is the "ratio of the number of commission payments to the number of transactions made" really a good parameter? And we are looking for a minimum value, as I understand it. And what is meant by commission? The brokerage companies for forex declare it virtually as zero, some brokerage companies "squeeze" the spread. Or is it just a spread? If we mean spread, it seems to be always available (it is said that there are some brokerage companies that trade without spread). Or does it mean the profit? But in this case its value is also questionable.
No. I meant any rebates to brokerage companies (spread, commission, swaps). Of course, they are present in each transaction, but the trick is that we may not close on each directional signal of the analytical block, thus the transactions become virtual and not charged with commission. Game on this.
Note, Sergei, we are not talking about a specific method of price series analysis and trading decision formation mechanism - it can be anything. This is an important point, because it allows us to formulate the general principle (appearance) of the optimal TS - it must be reversible and always "in the market".
Further we can talk about the optimal algorithm of forming the entry points and the optimal MM. Although the last problem has been successfully solved and moved from the field of fundamentals to the field of practical application. This combination: Algorithm-TC-MM optimized at each of its links based on the most general assumptions will provide us with the desired tool for working in the market. The best tool of all. We will no longer have to toss and turn at night and gnash our teeth at the thought that there may be somewhere better...
No. I was referring to any kickbacks to the DCs (spread, commission, swaps). Of course, they are present in each transaction, but the trick is that you can not close on each co-directional signal of the analytical unit, thus, as it were, the transactions become virtual and are not taxed by the DC commission. Game on this.
Note, Sergei, we are not talking about a specific method of price series analysis and trading decision formation mechanism - it can be anything. This is an important point, because it allows us to formulate the general principle (appearance) of the optimal TS - it must be reversible and always "in the market".
Further we can talk about the optimal algorithm of forming the entry points and the optimal MM. Although the last problem has been successfully solved and moved from the field of fundamentals to the field of practical application. This combination: Algorithm-TC-MM optimized at each of its links based on the most general assumptions will provide us with the desired tool for working in the market. The best tool of all. You won't have to toss and turn all night long, grind your teeth realizing that you may be somewhere better...
А! In that sense. Then I understand (as usual, it's all a matter of terminology). Roughly speaking, why should we close an order, if we are still in a trend, it is clear - we shouldn't. But it seems to me that the chosen prameter is not the most objective from this point of view, of course it is my imho, but it doesn't evaluate how well the strategy fits into a series of trends. But I could be wrong.
There is a slight subtlety.
The trend is over and the analytical unit generates a signal to close, and some time later generates a signal to open in the direction of an already closed position. This is the situation that has to be played out - to close and open at the same point in the price series.
Residuals are not cumm. the sum is the regression error (difference). And they are not necessarily white noise, nor are they necessarily stationary. But even if they are, it does not say that there are no dependencies left in the series. They can be as many as you want, and even deterministic. What matters in this aspect is the period of these dependencies - how long they last in time. The shorter their lifetime, the less they will stand out in the series making it non-stationary. In the limiting case, when they act during one reference period - their presence has no effect on the shape and parameters of the distribution. I.e., what matters here is the ratio of sampling rate to time of action of the dependencies.
This difference is not the first difference, but it is a big difference, so the residuals are already summed up by time and you may earn on it. The rest is right, I wanted to tell you about it. It's all about sampling frequency, let's say we get white noise from davids, we take ACF of this process and get minus 50%.Everything is logical, because the readings themselves tend to return. But the days consist of minutiae, transaction prices. If we take the ACF of the minutiae included in this process, there will be no dependence, and the process itself will be random within this process, but reaching the limit, it will go in the opposite direction also randomly.
Non-stationarity means that there are dependencies in the series that are sufficiently long in time (at a given sampling rate). But dependence is not necessarily a dependence on previous values of the series. Dependence on the starting point in time (as in the definition of stationarity and non-stationarity). For example, if trades turn out to be dependent, it doesn't mean that the market remembers their values or consistency but they just fell in time into a certain phase of the governing process development (up-trend, for example). I.e. dependence manifests itself by the fact that sometimes the series is distributed quite differently and this can last for quite a long time. And many people perceive it as in the case of dependent trades - that two losing trades will most probably be followed by a profitable one, though it is usually a random arrival at a certain phase of the managing process and not necessarily this time-synchronization with it will happen again in the future.
In fact, any TS looks for the moments when the distribution is stationary at the maximum and has a positive mo
I think you mean the dependence on the previous first difference, you're still guided by the price. You're writing about the local dependence on the starting point in time.
1. Too bad you don't understand the difference between the two statements: "the shape and parameters of the distribution do not unambiguously answer that there are no dependencies" and "the shape and parameters of the distribution indicate that there are local patterns". But that's your problem.
2. Well, since you said "A", keep saying it. Prove it.
3. This is an interesting situation. On the one hand you can prove that there are no local regularities on the random walk. This obviously implies that there are no global patterns at all. That is, there is definitely nothing to make money on. On the other hand, you are present on this forum and probably on forex. A logical question: what are you doing here? :-)
You need specifics, not cowardly shouting from around the corner at the right time.The question about local patterns in SB is left to you.
I'm sure the topic can now be closed for lack of practical usefulness of the issue under discussion.
You'll be second, after Reshetov. ;о)
Ну, или можно немного расслабится и слегка "пофлудить". Например, мне всегда было интересно построить адекватную модель ценового ВР. АР-модели для этого слабо подходили, т.к. не давали полной картины наблюдаемых явлений (например, отсутствовали "толстые хвосты" в распределениии РПР). Я в этой ветке, чуть выше, озвучил свою идею (намётки на идею) о возможной тонкой структуре ценового ряда. Основывается она на том факте, что на разных ТФ наблюдается одно значимое свойство - слабая корреляция (отрицательная кстати) между соседними отсчётами в РПР. Уже для отсчётов отстоящих друг от друга более, чем на один бар, значимой зависимости почти нет. Ключевой момент - зависимость эта есть для всех ТФ на выбранном ВР. Все АР-модели этот момент игнорируют! Они точно смоделируют зависимости между отсчётами в пределах конкретного ТФ, но стоит перейти на другой ТФ и всё, баста! - ничего по конкретной модели не работает.
Было бы интересно услышать мнение форумчан. Наличие адекватной модели ценообразования, позволит понять скрытые механизмы курсовой динамики, а значит есть не нулевая вероятность их профитной эксплуатации.
Strange man, you pay attention to such crap as ACF minus 2-3%, and ignore the dependence of 99.9%. Take WMT share,traded in all kitchens.ACF minus 10-20% with stat. significance, whether on days, whether on watch, I do not remember.Go and try to ruin even one DC.
By the way, not to be unsubstantiated. Here's this humble curve:
has the following characteristics
In addition it has an akf like this:
(Of course, it's not great, but it's quite workable)
Well, in addition, if we apply an inverse transformation we can get M15, EURUSD.
What do you mean, dependence on the previous first difference, you're still guided by the price anyway. You're writing about a local dependence on the starting point in time. I asked again, is it R/S analysis - timing, or dependence on a certain time of day (random EA with variable SL, TP, which opens positions at random, which found this dependence)?
I want to say that not necessarily the previous prices or their increments are the cause of future movements (though it's possible and appropriate TA methods use it, for example momentum trading). Sometimes price movements may be used to identify some market phase or its finish, which is exploited by pattern trading. It means that the prices or their increments only serve as a trace of continuation or termination of some objective market processes.
In both cases we have a price series dependence.