How can I tell the difference between a FOREX chart and a PRNG? - page 10
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
Dick, read the definition in the book, it's exactly the same as mine)) And you have a definition of martingale, to which some SBs can be reduced by a number of operations, but not always (in particular SBs can be Markovian, semi-Markovian and even non-Markovian, with any correlations and Green's functions, etc.)
Your definition is a muddled mess, not a definition. But it's the book's fault. And Markov of course))
(.... Thoughtfully so ....)
You, colleagues, can't figure this out without the nihilist Reshetov. The situation with "probability theory" is very bad and quite similar to DSP ("digital sound"). In the first one Kolmogorov intentionally screwed up, while in the second one our dear Vladimir Alexandrovich (Kotelnikov), from Kazan by the way, accidentally and unintentionally made a mess of it. And if we start to sort it out, (having Gödel theorem and Church's thesis in mind, and stuff like that), it becomes clear that ATTENTION, all these "statistical theorems" and "conclusions" are just tautology, having nothing to do with real problem solving. And without understanding this , we can swim in terms until we are blue in the face, without understanding each other.
Do you know who GeorgeMarsaglia is? If you're going to discuss "..... PRNG, you should know who he is. He made a couple of comments along the way, which for TRADING purposes put everything in its place (assuming that you know DSP as well as you know gpwr, or Privalov, or at least as the Hindu curator of L-SProgrammer).
Do you know who Professor Orlov is? In principle you don't need to know, just read carefully what Reshetov wrote here on the matter of probability.
Let me repeat, you Colleagues must first agree on the terms, and then argue. And here in the course of digging in terms, will reveal an appalling abyss of nonsense, on which the modern experts of "probability theory" leap. For example, there are no "Markovian processes". There are none, this is clear to anyone versed in physics, not just Lomonosov. Any system has PREMIERE, i.e. a history of development of states. And by the way your ability to predict just in the knowledge of this history, and not only "the current state". Strange that ALSU and GPWR (as well as some others on this forum), who I assume are GOOD at mathematics and DSP, and know that there are no filters without delay (without history), and here they cannot understand each other.
(... even more thoughtful.....)
Basically, if you ask each other questions here
"so what?"
"so what?"
and do it a LOT, literally after every sentence, and with a tone of actor Okhlobystin as doctor Bykov, you can come to nihilistic, but correct conclusions. The truth is that it will take a long time.
Personally I can, but I won't do it, sorry, I have a lot of work to do.
P.S. One of these days I will address some respected mathematicians here in a private matter with the results of my work.
P.P.S. Colleagues, you need to listen more carefully what you say to each other here. Some of the statements are simply invaluable. There are few mathematicians here. For example GPWR in his dialog skipped over valuable remarks of ASLU in other threads. And so on.
P.P.P.S. I apologize in advance if this post seems abstruse or arrogant to anyone. It is not.
I am surprised that you want to write huge posts about nothing when you have no time))
And I'm just giving beacons - where to look, and especially - WHERE NOT to look. And then, a person can be uninspired for 1-2 hours when tired, right? And what, I'm supposed to just yawn, looking at this skirmish, if I'm past it a long time ago? In principle, you don't have to "go through" anything, economists are taught "probability theory nihilism" at university.
Professor Orlov almost directly writes about this.
Excel is taken and a pseudo-random series is built using a function.
It includes trends, flats, movements in channels, false breakdowns, chart patterns, etc., etc.
How to distinguish real quotes from PRNG?
There is such a notion asa Wiener process and there is a filter that monitors this process. It is called a Wiener Filter
The technology is simple. We feed the analyzed process to the filter input and look at the output. If the filter jingles (jargon) then the analyzed process is not Wiener's, but different from it...then it's the turn of statistical radio engineering.... There are a lot of letters, who is interested, I hope he will follow the links and at least read them.
Z.I. We used to solve such problems with cadets at practical classes on radiolocation. The standard task is to distinguish noise at the radar input from a mixture of noise + signal...
Excel is taken and a pseudo-random series is built using a function.
It includes trends, flats, movements in channels, false breakdowns, chart patterns, etc., etc.
How to distinguish real quotes from PRNG?
May I also answer?
YOU CAN'T.
A pseudo-random number generator does not differ from a price quote in ANY way, except for length, period. This has long been known in trading. In Jack Schwager's book, a famous trader expressed this with a simple question, well known on Wall Street among analysts:
"How long is the coastline of England?"
The correct answer is "it is infinite". The more detailed you study the coastline, the longer the coastline will be. It's all down to the resolution of your instruments. This is "randomness" as in uncertainty. SProgrammer (his handler knows a thing or two about models and trading modelling problems) has said it here before, but everyone has let it slip past their ears.
Both have autocorrelation, only you can't always calculate it, that's where "randomness" comes in. So, as Eugene Slutsky's work well knows, if you are going to distinguish between a random variable and a random walk, the thorny question remains about the HAPPENESS of correlation between random variables. And that makes all the difference. Because if it is present, any correlated random variable will give, under sliding summation (or any filtering), a series with distinguishable sinusoidal "harmonics" of sorts. Which is what radio operators throw themselves at, and then don't understand why they are bouncing around instead of waltzing around like in radars.
Can I answer that too?
NONE.
"What is the length of the coast of England?"
Both have autocorrelation, only you can't always calculate it, which is where "randomness" comes in. So, as Eugene Slutsky's work well knows, if you take the distinction between a random variable and a random walk, then the thorny question remains about HOW much correlation there is between random variables. And that makes all the difference. Because if there is correlation, any correlated random variable will produce a sliding summation series with distinguishable sinusoidal "harmonics" of sorts. Which is what radio operators throw themselves at, and then don't understand why these are there jumping around frantically and not waltzing around like in the radar.
1. The coast of England - is this about the so-called fractality of the market?
2. What is the problem with calculating the autocorrelation of gpsc performance and market quotes? I have calculated the correlation in both cases. Some realisations are stronger than market quotes and some are weaker.
2. What is the problem with calculating the autocorrelation of gpsc performance and market quotes? I have calculated it - it is present in both. Some realisations are stronger than market quotes and some are weaker.
1. The coast of England - is this about the so-called fractality of the market?
2. What is the problem with calculating the autocorrelation of gpsc performance and market quotes?
1. Well in general, yes, you can call it "fractality".
2. No problem.
For that you must be well acquainted with works of Russian academicians in the field of radio engineering, with works of Slutsky in original, with works of scientists from Maxwell school about beginning of 1900th, more preferably with statements of Marsaglia, and something else. If you don't know that and start "calculating autocorrelation" with conventional or robust or non-parametric methods, you will get some results, but never the 0.1% accuracy required for retail forex.
1. Well in general, yes, you can call it 'fractality'.
2. No problem.
For that you should be well acquainted with works of Russian academicians in the field of radio engineering, with works of Slutsky in the original, with works of scientists from Maxwell school around the beginning of 1900th, more preferably with statements of Marsaglia, and something else. If you don't know that and start "calculating autocorrelation" with conventional or robust or non-parametric methods, you will get some results, but never the 0.1% accuracy required for retail forex.