Random Flow Theory and FOREX - page 53
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
A very interesting observation in a crisis.
there is an opportunity to make money in every market.
Well done! A hero! Champion! You're absolutely right. You've done quite a bit of work and you've broken through an open door. Naturally, the price increment is not a random walk. The price increment, usually in the form return = log P(t) - log P(t-1), is noise. With some assumptions it is assumed to be stationary. Again with some assumptions it is assumed to be normally distributed. If this assumption is too broad, it is narrowed down to a stable distribution, which has no solution in general form, only special cases. Random walk is the price itself.
If you have heard about martingale, it does not mean that all econometrics ends there. Every rule has exceptions, which other financial mathematicians have won Nobel prizes on. Consider it the ninth wonder of the world, I won't dissuade you.
Oh, I almost forgot - random walk is random walk, the definition is on wikipedia.
I see that you, Timbo, consider a reference to the Nobel Prize to be the pinnacle of mathematical proof. Ridiculous. Especially considering our discussion with you about those Nobel laureates who received their prize for proving the impossibility of crises in the modern financial and economic system of capitalism. Do you. by the way, do you know if they got their prize back or if they got squeezed, silently.
I would have accepted your references to Nobel laureates if you had seen their work with one eye or understood at least one line there. But since you haven't (and I have no doubt you have, since you have already proven that you have as little to do with financial mathematics as you do with psychology), you'd better not mention it.
And you're wrong about the open door thing too. You seem to think there's only one, which is the one you entered in first grade. No, it's not. There are many doors in maths, and most of them are closed. And what you've found on wikipedia or in your school books isn't a door, it's an alphabet.
I researched one-dimensional equal probability random walks. Exactly what is described in wikipedia. Not experimentally, as you might think, but theoretically. Obtained certain, absolutely rigorous mathematical results. Comparison of those results with "random price wandering" shows that it is definitely not random wandering. The reliability is 100%.
Hence the value of your babbling about "some assumptions". I count three. What does that tell you? You probably haven't even thought about it. And what it says is that all the models with which "financial mathematicians" like you crawl up to the price series are completely inadequate. That is, you can't refer to them, much less draw any conclusions about price behaviour based on them. Therefore, all your assertions are nothing but hogwash. You cannot prove any of them. You cannot even confirm them experimentally, at least partially.
That's why you don't need to tell tales here about noise either. The noise you so arrogantly raised here is enough already.
the opportunity to make money is in every market.
But not GER supporters. To rip off Pythians by telling them about Nobels is yes, but to drain your own deposit smacks of a psychiatrist.
But not GER supporters. To rile up Pythians by telling them about the Nobels, yes, but to drain your own deposit smacks of a psychiatrist.
The GER thesis that all information is in the price seems to be used by everyone who makes quotes series forecasts. Or are you doing something else?
Mass Service Theory. Every underground passenger randomly and independently arrives at the underground. For this random flow, it is possible to calculate both queue lengths and waiting times quite accurately. But this is just a matter of speaking.
As has been shown in a number of papers, capital markets are non-linear dynamic systems with feedbacks (a memory of about 40 months). Most academics, and more importantly Nobel laureates, argue that such a precise market model is unnecessary, and can be limited to the model known as the efficient market hypothesis. What is there to discuss here? Who has fallen from what oak tree? Any constructive discussion presupposes that existing opinions and approaches are taken into account. Otherwise, it is just flooding.
Careful with your expressions and emotions, otherwise it is unclear what you are arguing for and against. In the previous this post you seem to be against "Markov chains" (right?) - well, I agree, since no one has ever and nowhere seen an INDEPENDENT system state from anything. Markov chains and Markovian processes are blithering nonsense, and people write dissertations on them and defend them.
In this post about the underground - WARNING - you simply confirm Slutsky's idea that PARTLY random events, added together, can generate oscillatory-like signals. I note that in your example about the underground - the process of arriving is UNLIMITED, at least not as random as it is perceived in probability theory (since there will EXACTLY be no people in the underground tonight, and tomorrow at 08-00 there will be people going to work, so we know in advance that their arrival is UNLIMITED).
Thus, between the probability theory and the theory of mathematical statistics there is a deep contradiction in the applicability of definitions, but for some reason (unknown to me personally) they - mathematicians are not confused and does not prevent building air castles.
Random variables do not have "memory". All these statistical models don't work. None of these fucking "Nobel laureates" could foresee or calculate the 2008 global financial crisis. Not one of their lauded models has shown such a scenario (except for 5-9 second hand economists). So first you need to filter the definitions of the methods that are going to be used for trading, and only then shove all this to the MQL4 code.
People, please, for God's sake, give at least one link to serious literature, where mathematicians strictly prove the impossibility of winning in a game based on random walk, for example, on flipping the right coin.
After all, you do realise that everything depends on the conditions of the game. I recently gave just such an example, when the game conditions allow one of the participants to steadily beat the other exactly on the process of flipping a coin. It's true that creating the same conditions as in a game of Penny, when playing against the market, is unlikely to work.
But that is not proof that it is not possible to create a profitable system for a game based on a stationary random process. It is on broker conditions.
The difference between trading and "gambling" in the mathematical and gambling sense is the practical limitlessness of the resource with which the trader is betting. Moreover, the value of the stock-resource in the trader's pocket changes according to the course of life around him. Such conditions do not exist in any classic game. So mechanically transposing game theory to such a phenomenon as the global market is simply IMPOSSIBLE. Maybe some game theory methods can be applied to short-term trading, but no one knows or has ever studied which ones and how to determine the boundaries of those areas.
Falling fivefold is investing in high-risk assets. Probably, if shares had gone up fivefold, no one would complain. But these are two sides of the same coin. What we fought for, we got what we got. Normal markets do not fall like that.
The stationarity probably cannot be explained, because the random process of quotes flow is NOT stationary. Surely some of the shareholders know this.
If fat tails are concerned, one should forget about normal distribution, so such an assumption is inadmissible for the model used. There is a stable distribution for that. And it's not a trifle. The assumptions must be reasonable.
For that, that is what mathematics is developing a field called robust modelling, or robust statistics. But it is still a work in progress (i.e. it is just the beginning).
Why are you arguing so unproductively? So Timbo provokes, that's the kind of man he is. But why should the rest of us do it?
Arguing is supposed to be about truth, not about showing everyone that your opponent is wrong, but about helping him understand something. But you argue instead of explaining, proving, and not listening to each other. You need questions, you need clarification.
And to be completely consistent, forget about stationarity and approach capital markets in general and Forex in particular as non-linear dynamic systems and do not waste time on stationarity .
In practical terms, how? Do you have any insights?
For example, I have calculated a pattern assuming a stationary process. Now, how do I exploit this pattern assuming non-stationarity?
Do I need to investigate this regularity in a non-stationary process? And how can I do it without knowing the properties and nature of this non-stationarity?
To study non-stationarity by example of this pattern?
As a nerd to a PTU student: I don't believe in Fourier analysis for quote flows for certain reasons. I do not believe that quotes are white noise. If I have not made myself clear, accept a thousand apologies.
P.S.If you put a negro together from sinusoids, it's silly to claim he's not made up of them :) But actually, we're not interpolating here, we're extrapolating.
I'll twist what Avals just said: the real living Negro is not made up of sinusoids. He is composed of water, of cells, of blood and meat, he has a free will that guides him in all his bodily movements. Your image of a Negro can, in certain cases, be "decomposed into sinusoids", and even derive some benefit from it. For instance, you can use an image of a Negro running to continue the motion cycle of his legs and make a movie about the Negro's running. Only the real Negro will stop after 30 seconds, and your extrapolation of his image will show that he is running EVERYWHERE. And his mood has CHANGED! And he no longer wants to run! Therefore, the mechanical extrapolation of his running on the picture of the video will not do anything.
You don't have a Negro. You only have an image of him. That is why you will not be able to make long-term predictions by his image (by means of decomposition into sine waves). You have to know where he is running, why he is running and how many resources he has for running, when he will get tired, and how to determine from individual pictures that he has started to get tired.
Fierstein?
Reception.