How can I tell the difference between a FOREX chart and a PRNG? - page 26
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
Mm-hm. That's a depressing conclusion. To make it more cheerful, I will repeat it for you: George Marsaglia made some remarks that put everything in its place and show where the bridge is. (of course not directly, it takes a lot of thought, good knowledge of DSP and a hell of a lot of programming.) You can do it without Marsaglia, but it will be a much longer way.
Grandpa Marsaglia was kind of a nihilist, and as far as I understand, he even had a conflict with NIST - the monster of American scientific bureaucracy, which stupidly parasitized his works, and (attention) it seems they have flaws in standard encryption-decryption-hashing methods there.
For dessert (or as an appetizer for the kino-picture above) here's another simplest, but high quality RNG by Marsaglia (though they should be initiated by another RNG):
It couldn't be easier. Marsaglia's grandfather was good at maths, theorizing and statistics.
I have respect for all people, including the unfamiliar Marsalle, but I am very suspicious of any results.
I have a very specific task in front of me to analyse kotir. I know the toolkit for this analysis. The tools that are not applicable to kotir are discussed here on a regular basis. The main reason for the application of these inapplicable tools is the past training of the posters, which is not an argument for me at all.
Same Marsaglia. What problems has he solved in non-stationary series? Not known to me. The text of your code does not answer that question. Is the task of distinguishing a deterministic trend from a SB with drift (and therefore understand the HSPF algorithms)? I do not have such a problem in front of me. Nosko has seen such a problem and I haven't. I bluntly take a package and try to apply it to enrichment. I see two problems. The first. Not enough brains of my own. The second. There are a bunch of problems that are not being solved at the moment.
The topic of this thread is quite practical. There is a problem, but in my opinion, we should not deal with it: we detrend kotir, and what is the trend there, we do not get into it. After detrending there are a lot of problems that obviously affect the success of TC. So they should be dealt with.
1. I respect all people, including the unfamiliar Marsalla, but I am very suspicious of any results.
2. I have a very specific task in front of me to analyse the quotient.
3. I know the toolkit for this analysis.
4. Tools which are not applicable to quotient are discussed here on a regular basis.
5. The main reason for using these inapplicable tools is the posters' past training, which to me is not an argument at all.
6. Same Marsaglia. What problems has he solved in non-stationary series?
7. Not known to me.
8. The text of your code does not answer this question.
9. Is the task to distinguish a deterministic trend from a SB with drift (and to understand the HSPC algorithms in this regard)?
10. I have no such task in front of me.
11. Nosko has seen such a problem, I have not.
12. I bluntly take the package and try to apply it to enrichment.
13. I see two problems. One. Not enough brains of their own. The second. There are a bunch of problems that are not being solved at the moment.
14. The topic of the thread is quite practical.
15. There is a problem, but, in my opinion, it should not be dealt with: we take and detrend kotir, and what trend there is, we do not go into it.
16. After detrending, there are plenty of problems that clearly affect the success of the TC. That's what we have to deal with.
1. Right. I agree.
2. Why do you need this analysis? The answer is partially in your paragraph 12. Don't make me laugh at this question, answer yourself first. The question has a perfectly mathematical reasoning.
3. Let's say. So what? (I roll into the image of Doctor Bykov for simplicity). Suppose they didn't find the formula "in the bag", or what they did find doesn't work, doesn't give a prediction.
4. Exactly, I don't have to tell you how to do it right either, it's my results, my time, and finally my money. But I have the right to show you WHERE I don't have to. Here, I show you.
5. Right. I agree.
6. Colleague, get out of this whirlpool called "stationarity". God, how many times can I repeat, suggest, ask - try to write a chain of definitions BACK
"stationary series" - "mat expectation" - "average" - "what is average" - "unchanging average" -..... OPINION - again (!) "stationary series" . This is a tautology.
And then, those who are thought to have "decided" - Hatanaka, Gringer, others - are they all WHERE? And if they are supposedly "solved" - what are we all doing here? Take their clever formulas from 600+ page books - and shovel money in.
7. So what? What conclusion follows from the fact that you don't "know" anything there?
8. It's answering, it's answering. You just don't see it, or you're not ready to see it. It's written in plain English. You just have to know it clearly by heart, be able to substitute ALL mathematical definitions relevant to the subject matter of this forum thread from your head on the fly. Moreover, there were probabilistic mathematicians between Carl Pearson and Kolmogorov who went about it the right way, but history has shuffled things around and Kolmogorov's wow followers have mixed things up.
9. What is a "deterministic trend"? What is "deterministic" and what is "trending"? Don't tell me, "Oh, you don't even know that?" I do know - it's you swimming from one definition to another, not giving yourself the trouble to double-check whether the definitions coincide with the theories and conclusions. You say (well, you don't say, but you think (C) Baron Munchausen) that you have a definition of "trend" somewhere. But you do not. A trend is just a special result of de-trending by ONE of the METHODS. Which one? Are they all the same? So all the "trends" will be DIFFERENT. So how can one define or at least just use the concept of "trend" to determine "stationarity" or something else?
10. WHAT is the problem? Formulate it mathematically, strong woo ple. How can you mathematically define the task of trading?
11. Possibly.
What does "apply" here? Explain in detail, in general terms, BUT BELIEVE the mathematical terminology.
13. So what? All scientists have faced this problem. This is called "science".
14. Exactly. Not dabbling in "stationarity" and "detrending", which are privately defined through each other. That is not science.
15. Why, is it? Why "don't we get into it"? Where is it written that you don't have to "get into it"? Well, I'm getting into it.
16. Or maybe it is not necessary to "detrend" globally? Doesn't the trend change, is it constant as it is stated in all theoretical formulas?
.............................................
Here I have answered briefly and clearly. Sometimes, the right question solves half the problem.
For several pages I try to follow this verbiage - I can't understand anything.
Prival says that there is no other way to calculate autocorrelation, no matter how much one criticizes the existing one, and in response the reference to Moses and Aaron.
The thesis that in reality it is almost impossible to distinguish real quotes from gpsch, and in response the remark that George Marsaglia showed several bridges. And in addition another gpsch is given. why?
That is a lot of bukouffe, a lot! Please make it short and to the point, otherwise people will get scared.
I can do that. That's easy.
How much?!
Here's my answer short and to the point. Sometimes the right question is half the battle.
475 words, 19 question sentences, not a word of substance - short and to the point.
I should have left out of all this nonsense only this"Exactly, I don't have to tell you how to do it right either, it's my results, my time, and finally my money.". Everything else is pure stream-of-consciousness thinking.
Write shorter and more to the point, or people will get frightened.
475 words, 19 question sentences, not a word of substance - short and to the point.
I should have left out of all this nonsense only this"Exactly, I don't have to tell you how to do it right either, it's my results, my time, and finally my money.". Everything else is pure stream-of-consciousness unclouded by thought.
Ah, yes, what am I doing? I have joined with my incomprehensible questions and allusions here on a forum, where others, unlike me, always express themselves exclusively for the cause, always express themselves exclusively clearly for others, and exceptionally productively and effectively for themselves and for others, and most importantly - always exclusively mathematically verified. I am horrified to realise my mistake and leave.
See you in a year. Speaking here once a year is enough. If anything, I will write to the knowledgeable mathematicians of this forum in private.
With horror I realise my mistake and walk away.
Here is my answer in short and to the point. Sometimes, the right question is half the battle.
Your post is much broader than that. So I won't comment on the whole thing.
Detrending. It comes in all shapes and sizes. From linear regression in the terminal to wavelets. You always need the purpose of such an operation. The topic is more than interesting, but goes beyond the scope of the topic. Open the topic, I'll be sure to participate in it.
For the purposes of this topic, I argue that distinguishing a deterministic trend from a stochastic trend is nonsensical. The reason is as follows. I believe that stochastic trends are an invention of dissertation students in statistical departments, far removed from concrete economics, as the economy is not based on GHRP, but on very deterministic and inertial processes - the production of goods and services. The stochastic nature of the quotes we observe is a froth on said deterministic processes. I am not arguing that froth cannot have a prevailing character, but I believe that the base of economic series - the production of goods and services - will far more often produce deterministic trends, rather than the stochastic ones that I have neglected.
Quite specifically on the topic of the thread. No need to try to recognise SB with demolition. That is my opinion. The author of the thread has a different opinion.
Privalov, you'd better correct your ACF and explanations to it, you'd brush it up, figuratively speaking. Of course, even such an ACF in the codebase is a hundred times better than no ACF at all. But still, this is a new thing, people will want to understand. And to avoid repeated bloody cuts, like the one involving nihilist hrenfx
"zero sample correlation does not mean there is no linear (usually forget the word linear too) relationship in that sample."
https://www.mql5.com/ru/forum/128968
in which 5-6 knowledgeable mathematicians on this forum never understood each other - so for that you should put beacons: where is the theoretical auto- and simple correlation and where is the sample correlation, how do the sample sizes and auto-correlation plot relate, how are the multipliers normalized. Otherwise it turns out that your periodic pure sine function has a damped autocorrelation
https://www.mql5.com/ru/forum/128968/page15
The autocorrelation formula given above in the wiki-Pedo-Wiki is understandable and suitable for pogramming, while yours is not yet.
This is not my intention to criticise, it's for clarification.
Regarding everything else, there are several robust and non-parametric ways to calculate autocorrelation, but here we ("no, now it's you"(c)) are entering the thin ice of the connection between theorwave and DSP, and personally I don't want and can't go further.
"2. There is no other formula, i.e. no matter how much we discuss calculating autocorrelation differently (no one does). "
Privalov, be careful with negative statements. Firstly, negative statements themselves are difficult to prove - in mathematics, mathematical logic, philosophy, orthodox theology and even Moses and Aaron's Judaism.
Secondly, if someone claims that chocolate cakes do NOT revolve around Saturn, how can this be proven and verified?
Thirdly, how do you know there is no other formula? You don't have to answer, it's a rhetorical question.
"But the time came andSlukin Gennady Petrovich showed the solution, just think of the beauty of this phrase he found the solution"With sufficient accuracy for practice..." "
Well, he did SHOW that accuracy, in percentages. That is, you know in advance approximately how accurate his method will be. But the author himself and his friend Yul can't show any precision in the correlation formula. Because it will go anywhere - depending on the shape of the distribution. No one makes instruments with "technical accuracy", since all instruments (and all mathematical methods) have a pre-determined PERCENTIALITY (residual term, magnitude of smallness, etc.) to work with. This is what they do in both engineering and science.
I have shown and told you quite enough on the subject, so hereby I bid you farewell.
It is not my formula. Don't attribute it to me. I got it from textbooks and maths packages. I didn't make it up. It's exactly the same on the wiki. The formula matches 100%. What's to clean up?
I andhrenfx explained that one should not confuse QC and ACF are different things. Even at a primitive level QC is a number, ACF is a function (many numbers).
You cited my picture, where I tried to showhrenfx the difference.
Otherwise it turns out that your periodic pure sine function has a damped autocorrelation...."
Yes it does, and I want to point out it's not me. And MathCAd, add here and MathLab in it exactly as it turns out, because lcorr(Y,Y) - this function is built into matcad, I have not programmed it and did not invent ... (Anyone who knows matcad can go check it out) Do you honestly believe that both of these math packages don't calculate ACF correctly?
As for everything else, there are several robust and non-parametric ways of calculating autocorrelation,
please give me the formula. I would very much like to see a robust one, and a non-parametric one as well...
I know very well that proving a negative result is very difficult. But it is not our case. Because this case is similar to the following one. c^2=a^2+b^2. Everyone knows this formula, Pythagoras' theorem. But suddenly someone comes along who says it can be calculated differently. Great, I don't mind, show me the formula, how would it differ from the above ?
yes exactly so it turns out, and I want to note not at me. And MathCAd, add MathLab in it exactly the same way because lcorr(Y,Y) is a function built into matcad, I did not program it and did not invent ... (anyone who knows matcad can go check it out) Do you honestly believe that both of these math packages don't calculate ACF correctly?
Why argue about who is cooler, the explanation is simple enough - the original signal is a segment of a sine wave in a rectangular window, its ACF is also a segment of a sine wave, but in a triangular window, i.e. exactly what we see in the second figure. This can be checked by elementary calculations. If we take a sinusoid unbounded in time, its ACF will be the same sinusoid. Conclusion 1: The mathdeck calculation is correct. Conclusion 2: if we calculate in this way the sample ACF (and not the actual ACF, which we will never know) of the real signal, we must not forget that the calculation is done in the window and, therefore, the result is always distorted.
QC and ACF are not to be confused. Even at a primitive level, AFC is a number, ACF is a function (many numbers).
With all due respect, ACF is defined as the dependence of QC on the distance between counts, so the difference is not so fundamental. And the classical formula itself (which, as rightly pointed out above, does imply at least stationarity of the process in the narrow sense, plus its ergodicity) confirms this.