Filter without delay - page 11
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
А Вы уже научились инвертировать массивы double? )))) Вопрос к программистам-изменение знака элементов большого массива
Даю наколку для формата double: 63-й старший бит - это знак. Надеюсь Вы не на MQL программы пишете?
The nakolka is cool. Has Vdev forgotten what inverse and auxiliary code are - forms of representation of negative numbers? What about eight years of DSP experience?VDev has forgotten what inverse and complementary code are - forms of representation of negative numbers?
This is used for integers. For real numbers in "IEEE number something or other" format, the sign bit is separate. So VDev makes a good point.
I've been working with DSP and DSP for 8 years now, but I'm not proud of it, and I wish the same for all of us.
In fact, the only way to find out who's a real badass and who is just showing off, is to monitor onyx )).
Don't you agree?
Screw coolness. I agree to be the uncoolest person on the forum.
For almost a year I've been trying to prove the futility of any TC associated with the word DSP, normal law, Fourier, etc. The reason is banal: it's not on the market. The marketplace itself is a different object, not reducible to a stationary process. It certainly has stationary plots, but IMHO we should deal with the object we have, not change the object to our knowledge.
This applies to almost everyone on this forum. Once upon a time the biggest local authority wrote "waillet is still a scam". As always without argument, but that's what he is an authority, he knows all the behemoths on the forum very well.
Rinquet is not a stationary process. More than that: an uncertain process, i.e. events that affect quotes that cannot be predicted (Iraq war).
Most TS do not last long and the reason is the same - the market has changed, the periodicity of the market has changed (the period of the dummy or any other indicator has changed). No one knows how to catch the moment of this change. It is possible to bring the TS to profitability due to optimization (branded completely), but there is no guarantee that while you optimize, the market will not change again.
It is similar to frequency modulation, but there is a law there which is recognisable. In the marketplace the frequency changes, but I haven't heard of anyone managing to adjust to that.
There is no need to fear the matlab itself. Rashid did part of the work (he doesn't know about it) - he described the output from MQL5 to DLL in C++. Matlab itself has C++ and builds DLLs. Having this transition, all functions of Matlab become available. And there we will be able to get a number, if not a diminutive one. There is huge domestic literature on wavelets, but it's in geophysics and cardiology.
Screw coolness. I agree to be the uncoolest on the forum.
For almost a year I've been trying to prove the futility of any TC associated with the word DSP, normal law, Fourier, etc. The reason is banal: it's not on the market. The marketplace itself is a different object, not reducible to a stationary process. It certainly has stationary plots, but IMHO we should deal with the object we have, not change the object to our knowledge.
This applies to almost everyone on this forum. Once upon a time the biggest local authority wrote "waillet is still a scam". As always without argument, but that's why he's an authority, he knows all the behemoths on the forum very well.
Rinquet is not a stationary process. More than that: an uncertain process, i.e. events that affect quotes that cannot be predicted (Iraq war).
Most TS do not last long and the reason is the same - the market has changed, the periodicity of the market has changed (the period of the dummy or any other indicator has changed). No one knows how to catch the moment of this change. It is possible to lead to the profitability of TS by optimizing (branded completely), but there is no guarantee that while you optimize, the market will not change again.
It is similar to frequency modulation, but there is a law there which is recognisable. In the marketplace the frequency changes, but I haven't heard of anyone managing to adjust to that.
There is no need to fear the matlab itself. Rashid did part of the work (he doesn't know about it) - he described the output from MQL5 to DLL in C++. Matlab itself has C++ and builds DLLs. Having this transition, all functions of Matlab become available. And there we will be able to get a number, if not a diminutive one. There's huge domestic literature on wavelets, but it's in geophysics and cardiology.
And how else can you make money if you don't use plots with a stationary distribution? After all, from the entry point, to the exit point, there is a stationary distribution with positive mo. Another thing is whether the strategy for finding entry/exit points will be constant, or whether it will be constantly adapting to market changes. But the adaptation can be different. This is either a known correspondence in advance: parameter of market changes -> parameters of the system, or this correspondence is unknown in advance and is searched for every time anew. In the last case the question of statistical validity will be pressing. imho.
How else can you make money if you don't use areas with a stationary distribution? After all, from the entry point to the exit point, there will be a stationary distribution with positive mo. Another thing is whether the strategy for finding entry/exit points will be the same, or constantly adapt to changes in the market. But the adaptation can be different. This is either a known correspondence in advance: parameter of market changes -> parameters of the system, or this correspondence is unknown in advance and is searched for every time anew. In the latter case the question of statistical reliability will be pressing. imho.
What does stationarity have to do with it? The trend is going on and it is OK, it does not matter what kind of trend it is. I saw a paper where it is proved that there are segments of BP when one cannot say whether it is stationary or not.
I would generalize TA in the following way: search for a pattern, check it, get it, go ahead. It's a principle like Chukchi - what I see, what I sing. There's a new breed of Chukchi - NS: they sing what they can't see. But everyone relies on the TA law "history repeats itself". I've seen quite a few examples of adaptation, but never understood what one is adapting to. It all has one thing in common - it dies over time.
In all TA there is no model rinket. On the forum it is believed that something can be done by either counting or bringing the market to a stationary one. But it is not stationary. There is a tool plus the ignorance and laziness of those gathered that prevents mastering the waylets in matlab. It's a bit much for one person, but quite manageable for a few. Some geophysicists, cardiologists have mastered it.
Почти год я пытаюсь доказать бесперспективность любых ТС, связанный со словом ЦОС, нормальный закон, Фурье и т.д. Причина банальна: этого нет на рынкете. Сам рынкет - это другой объект, не сводимый к стационарному процессу. У него конечно имеются стационарные участки, но ИМХО нада заниматься тем объектом, который имеем, а не менять объект к своим знаниям.
Это относится практически ко всем участникам форума. Когда-то самый крупный местный авторитет написал "вейлет - это еще тот лохотрон". Как всегда без аргументации, но на то он и авторитет, что хорошо знает всех бегемотов на форуме.
Это похоже на частотную модуляцию, но там имеется закон, который можно распознать. На рынкете частота меняется, но я не слышал чтобы кто-то сумел подстроиться под это. (****)
:) For DSP, well done. For wavelets, that's... :) How can I put it to you, why not use Shannon coding? You need wavelets where you need them. And for those purposes and tasks for which they are invented. But I'm sure you don't have such purposes and tasks. :)
:) Well, as for DSP, IMHO your argumentation is correct for me, and in general we have essentially the same DSP. :) And "fight" it is necessary not with the DSP and with some techniques which are there by the ears "because it is lighter".
:) For DSP, well done. For wavelets, that's... :) How can I put it to you, why not use Shannon coding? You need wavelets where you need them. And for those purposes and tasks for which they are invented. But I'm sure you don't have such purposes and tasks. :)
:) Well, as for the DSP, your argumentation is correct for me, and in general we have essentially the DSP all the same. :) And "fight" it is necessary not with DSP, and with some techniques which are there by the ears "because it is lighter".
I can't say anything about Shannon, although it is a signal coding system.
Waylet is attractive because it is the same as a Fourier filter, but has a time duration. This agrees very well with the physics of the market: a team of bulls gets together and goes up, they cut their money and run away. This is the basis of the market and nothing can be done about it.The ratio of these teams we see in the form of kotier. Weylet supposedly allows you to determine the timing of the team's build and its breakdown, and provides a filter for selecting the strongest team from the rest of the crowd. In addition to this, the weylet has a predictive function. What I particularly like about veylet is that it analyzes the market for what it is. The forum is constantly trying to analyze what they taught in kindergarten.
Причем здесь стационарность? Идет тренд и ладно, а какой он там, не важно. Я видел работуЮ в которой доказывается, что бывают участки ВР, когда вообще невозможно сказать: стационарный он или нет.
Я бы обощил ТА следующим образом: ищется паттерн, проверяется, прибыблен - вперед. Это принцип как у чукчей - что вижу, то пою. Есть новая порода чукчей - это НС: поют, что не видят. Но все опираются на закон ТА "история повторяется". Я видел достаточно много примеров адаптпции, но ни разу не понял, к чему адаптируется. Все это обладает одним свойством - со временем умирает.
Во всеи ТА нет модел ринкета. На форуме считается, что можно что-то сделать, если либо считать, либо привести рынкет к стационарному. Но он не стационарный. Есть инструмент плюс невежество и лень собравшихся, которая не позволяет освоить вейлеты в матлабе. Для одного многовато, но для нескольких человек вполне подъемно. Освоили какие-то геофизики, кардиологи.
I am personally close to your perception. By the way. :)
Are you even a programmer? I have one "idea", not even an idea, but an interesting "task".
I myself now do not have time for it, time that I have must be spent on other things.
I thought - what if I propose it to a couple of PROGRAMMISTS from this forum. But with a condition to make the results publicly available.
I'm offering it to you because I see that your viewpoint coincides with mine in this matter.
I can write to you in a private message with this proposal. Type of I told you my "task" and you would gather a few more people that would realize it.
I'll tell you right away it's not a CU. This is more of a tool.
Про Шеннона не могу сказать, хотя это система кодирования сигнала.
Вейлет привлекает тем, что это такой же фильтр, как по Фурье, но имеет длительность по времени. Это очень хорошо согласуется с физикой рынкета: собралась команда быков и поехали вверх, нарубили бабок - и разбежались. Это основа рынкета и сделать с этим ничего нельзя.Соотношение этих команд мы видем в виде котира. Вейлет якобы, позволяет определить время сбора команды и ее развала, а попутно дать фильтр для выделения наиболее сильной команды среди остальной толпы. Кроме этого, вейлет обладает прогнозной функцией. Особенно меня привлекает в вейлете, что он анализирует рынкет таким каков он есть. На форуме постоянно пытаются анализировать то, чему научили в детском садике.
It "predicts" the market in inverted commas in about the same way as Fourier. That is, simply by copying.
There is a forex market and a strawberry market, so even if they both have the word market and the charts are externally similar, they are completely DIFFERENT on the inside. What I mean is that you are talking about a bull team.