Spectrum derivative (or spectrum acceleration) - 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
Well . since the issue of ticks has been raised here. I will repeat and then I will continue later in another thread (there are many such threads in the forum ).
So hypothetically imagine that we have a data stream signal + noise (but don't say that there is no noise in the market, it is clear, because we distinguish the useful part on the background of the whole list of data, so this useful part depends on data completeness). We know both signal and noise. As the noise component is large, then let's assume such a scheme that the signal is buried in noise and can not be discerned, but in some places it is still above the noise.
At least partially we can judge about the signal by such moments.
Further I will give you an example on the picture, but don't criticize it, my hypothesis is about getting into ticks or not.
1 - On the one hand it is believed that the ticks can improve the accuracy, but you can do without them, ie roughly speaking the signal is still recognizable in older ff. In this case we may forget about ticks and use bars (I have some doubts, because I have not checked them all yet).
2 - But I have trouble with the following moment. What if the signal can be recognized only at the lowest sampling rate (ignore the multicurrency analysis or analysis one pair) in principle, what if the signal can be recognized only at certain sampling rates and changes in phase, above which the signal is simply buried and it is not possible to recognize. The picture shows the evolution of the process at the lowest sampling rate (picture A) and with increasing sampling rate (picture B) - the red is the signal and the black is the signal+noise (not very good pictures, but the idea is clear). It is obvious that the signal disappears without any possibility to recover it even with small interferences. Everything is disturbed for sure and everything flies to the furnace (fuck and my hope for an easy life without ticks), if the Kotelnikov theorem is violated + moreover both period and frequency are floating.
All of this makes me very wary. Maybe I'm pissing against the wind, and analyzing minutes as the smallest required sample rate I no longer have to recognize the signal.
In markets where the tick history is the same for all - futures, there are always arbitrage situations, so there are reasons for them, so it makes sense to look for a signal, and the ping rate is a very good thing. But in this era of technology, such situations are resolved in no time, so the threshold of detection is nearly one point (or maybe several points), so think about what we are trying to get out of the minutes. I read Prival's posts and became even more hesitant to find similar thoughts in a person who is already versed in DSP.
And he was talking about H-volatility and its threshold, and about Kotelnikov's theorem. In some thread troll 222 wrote about his posts.
But nobody discussed it. I would like to discuss it. Anyway, I am too lazy to look for it, I will bring up that thread later.
It cannot be otherwise, arbitrage situations will not disappear, as generated by the very essence of the market, therefore, noise and struggle for speed of information may only increase, it is quite sad in this light. This is what I have in my head, and how to say it mathematically here I will think about it further
------------------------------------------------------------------------------------------------------------------------
I'll sum it up a bit. In the end Pryval is right saying, that we need the system of stochastic difurcations, but nevertheless this system is only useful for ticks analysis or for minute analysis, but with preliminary ticks filtration. Hence weighting coefficients (as I noticed) - in fact they can't be taken without ticks and they tell only about the level (amount of tick signal roughly speaking) of signal and hence the influence in combination with other pairs (tick volumes are different, and the signal level after tick filtration will be different). The beginning of changes of different pairs of coefficients will be different, something (no matter what) will go in front of it, and then something else will go in front of it. When will I get to that? Besides, spread may play decisive role in signal detection.
-----------------------------------------------------------------------------------------------------------------------
If we want to analyze bars then only phase analysis is needed, it is already the consequence of tick "root causes" and it only makes sense to analyze the superposition of common frequencies and common cycles (perhaps to see how the common frequency beam is scattered and narrowed), but since the phase, the amplitude component in tools different in amplitude characteristics is different, therefore it is necessary to get rid of it. And I forgot, if all floats, you have to take all frequencies with no exceptions, you can not throw anything away, after mixing them will show which are taken and which will be discarded.
Not just words to say it all, and mathematically not every advanced mathematician would even get (not to mention myself) how to express it, that's the tricky part. And you are all so vague and muddy, if I could just put my brain out like that once and put it all out.... Ann no, the hell with it , every time I try to explain it, I get some ideas about how to formalize it, so it's not worth it to stop.
It's not the smell of green that captivates me,
It's not finding a better angle,
It's the chance to stay true to myself
And not to depend on dough. (Alexander Sergeyevich Trollolo. To be continued......))))
Who's insisting, you asked me where I shared, I showed you.
There are many ways to score. Sometimes it makes someone feel bad. Maybe you should learn how to phrase it properly. Then there will be no need to score. (Just a tip, nothing more)
Scoring can be done in different ways. Sometimes it makes someone feel bad. Maybe you should learn to phrase it properly. Then you would not have to hammer. (Just an advice, nothing more)
I have to write explanations at every step of calculations, and the file is elementary. I just wanted to say that sometimes frequencies may override each other in importance (lower frequencies may sometimes take away the importance of higher frequencies when defining the signal).
we compare how, take a cut of one frequency from different instruments and compare or do something else with it .... And I meant that it would be nice to consider a compound frequency (sounds wild) in which a mixture of periods of different frequencies. I haven't figured it out yet.
I have to write explanations at each step of the calculation, and the file is elementary. I just wanted to say that sometimes frequencies can override each other in terms of importance.
And who knew about it at the beginning. Only at the end does it become known. If it had been written right away, everything could have gone in a different direction.
Although most often the importance is always determined afterwards, at the beginning it is never known (this is in your defense).
And who knew at the beginning. Only at the end does it become known. If you had written right away, things might have gone in a different direction.
Although most often the importance is always determined afterwards, at the beginning it is never known (this is in your defense)
that's what i'm talking about, they ask to explain something understandable, it's like asking to lay out the calculations of an already made grail)))) but if i don't have these calculations how can i express it (search directions), so i try to tell little by little what and where i dig. for many it is difficult to understand because people think in images, but to understand it at the level of imaginary thinking i need a complete breakdown of everything in it, it is difficult to do for the visual analysis. It is like seeing the deep diffusion process of two substances for each molecule at the same time, vision can switch from one molecule to another, analyzing them one by one in relation to other molecules, and then merging the results of analysis into one.
That is why computers should do it, and give ready results, simplified for visual analysis, when all information, compressed and transformed for understanding by vision and mind, is in front of the trader.
I still don't understand what you mean by phase, trollolo. You talk about it, you talk about it, but you don't give any definition.
What is a phase?
that's what i'm talking about, they ask for explanations, it's like asking for calculations of the ready grail )))) but if i don't have these calculations how can i express (search directions), so i try to tell them little by little what and where i dig. for many it is difficult to understand because people think in images, but to understand it at the level of imaginary thinking i need a complete breakdown of everything in sight, it is difficult for visual analysis. It is like seeing how deep the process of diffusion of two substances for, say, each individual molecule is, the vision can switch from molecule to molecule, analyzing them one by one in relation to other molecules, and then merging the results of analysis into one whole.
The information that has been transformed and compressed for comprehension by vision and mind is in front of the trader's eyes and mind, but it cannot simultaneously monitor everything and analyse several processes at the same time.
That's what I'm saying, it's hard to talk about things you don't understand. First you have to understand it yourself, then explain it to others. Otherwise they don't understand what it's all about the first time. Maybe you should look for a higher level forum. This one is wrong. If you tell them once, they don't understand. You tell them twice, they don't get it. If you tell them three times, they don't understand. You start to get it yourself, but they still don't get it. It's hard, of course. I understand and sympathize.
And it is necessary to analyze several processes at a time. Otherwise you may have to choose between trading and forum.
I apologize if I have said something wrong. I did not want to offend anyone. Only sympathy and nothing more.
I am saying that it is difficult to talk about things that are not clear. First you have to understand it yourself, then explain it to others. Otherwise they don't understand what you're talking about the first time. Maybe you should look for a higher level forum. This one is wrong. If you tell them once, they don't understand. You tell them twice, they don't get it. If you tell them three times, they don't understand. You start to get it yourself, but they still don't get it. It's hard, of course. I understand and sympathize.
And you have to analyse several processes at the same time. Otherwise you will have to choose either trade or the forum.
I apologize if I said something wrong. I did not want to offend anyone. Only sympathy and nothing more.
It's OK, it'll get better and clearer, the main thing is not to stop, thanks to the Mathcad.
Lots of letters. All incomprehensible. I didn't expect more from you, and I didn't count on it.)
I did not expect more from you, and do not expect)))) you have already decided that the topic is not interesting to you, why are you drawn here? If you laugh, I understand, myself without humor, but remove the trash then for the failure.