Statistics as a way of looking into the future! - page 13
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
On a nutshell: the market is a complex dynamic system. What more do you need? :)
>> yeah, or a non-complex dynamic system. :)
We need, the underlying premise, why does a theory that is supposed to unambiguously determine the future of a system by implicit but deterministic laws, suddenly become suitable for predicting future prices?
For a start - do we need determination?
In general, for me personally, the theory of dynamic systems has the same relevance to the market as Mendel's laws. No more than that.
Don't misunderstand me, I'm not asking about the name of the "curve mare" (the Theory of Dynamic Systems) you're going to ride around the market with, but about why exactly did you decide that the Theory of Dynamic Systems is saving? If the Dynamic Systems Theory is something familiar to you, then you can easily explain on your fingers the philosophy and the essence of how the Dynamic Systems Theory uncovers the market and predicts the future. For me personally, the mechanistic transfer of Dynamic Systems Theory to the market is unacceptable just because the theory contains the words "predict the future behaviour of the system". What makes you think that the market will fall under the Dynamic Systems Theory?
May I try to answer. Maybe it's not very correct to answer a question with a question, but what theory does it (the market) follow?
I believe that a good TS cannot be built without making a forecast, let it be 0.62 probability, it means that I enter the market with SL=TR in 62 trades out of 100 and get a guaranteed profit.
Who tries to make a forecast using regression, someone uses multivariate analysis, someone trains NA without knowing what for, hopefully it will succeed (like NA is smart and will answer all questions by itself), many people build various TFs using harmonic expansion, and I, for example, stuck to the system of stochastic dif. I chose the system of stochastic differential equations only because it worked, not for Forex, but we are shooting down aeroplanes and they fly too, though they are heavier than air.
You cannot do without a forecast, otherwise it's like jumping in with your head in a cloud. Many have already taken to the water, millions have drowned, many will not come back, there are those who believe they can forecast and are looking for it. Some have found it (or think they have found it), and they either return here to search for it, or leave the forum (as they are not interested). There is no third.
Please explain the meaning of "one of the factors influencing gold is the dollar". The price of gold is usually taken in relation to the dollar, i.e. all the gold-dollar influencing factors are already taken into account. Or should we use some kind of dollar index that excludes the correlation to gold?
You have answered the question yourself. If you know English go to http://www.gold.org/, or rather http://www.invest.gold.org/sites/en/why_gold/gold_and_the_dollar/ there is a topic on the subject
May I try to answer. Maybe it's not very correct to answer a question with a question, but what theory does it (the market) follow?
I believe that a good TS cannot be built without making a forecast, let it be 0.62 probability, it means that I enter the market with SL=TR in 62 trades out of 100 and get a guaranteed profit.
Who tries to make a forecast using regression, someone uses multivariate analysis, someone trains NA without knowing what for, hopefully it will succeed (like NA is smart and will answer all questions by itself), many people build various TFs using harmonic expansion, and I, for example, stuck to the system of stochastic dif. I chose the system of stochastic differential equations only because it worked, not for Forex, but we are shooting down aeroplanes and they fly too, though they are heavier than air.
You cannot do without a forecast, otherwise it's like jumping in with your head in a cloud. Many have already taken to the water, millions have drowned, many will not come back, there are those who believe they can forecast and are looking for it. Some have found it (or think they have found it), and they either return here to search for it, or leave the forum (as they are not interested). There is no third.
Of course it's OK, the question is very valid. The market will fall under a theory that makes no unacceptable assumptions. :) In my view, the Dynamic Systems Theory voiced here is unsuitable precisely because the assumptions necessary to stack the market in it are not at all acceptable.
If you see, I ask to point out exactly those assumptions about the market, which must be made before one can lie in the bed of this or that theory. Experts in dynamical systems theory, regressions, multivariate analyses, etc. should know this and be able to tell you about it.
As for the forecast, without which there is no way, it all depends. The first thing that comes to mind is forecasting the price. That's what this thread is all about. This is an obviously futile way, in my opinion, because there are not enough confirmed assumptions about the market to make any forecast, except for the average conceit.
About the planes and the process of throwing into the whirlpool I completely agree. But process philosophy says that if "profit" disappears from a forecast (physicists once had matter "disappear", just because their ideas about matter were wrong), then I suppose our forecast's ideas about profit (the market) are wrong. Sort of.
As for the forecast, without which there is no way, it all depends. The first thing that comes to mind is forecasting the price. That's what this thread is all about. This is an obviously futile path, in my opinion, because there are not enough confirmed assumptions about the market to make any predictions other than average conceit.
I take it you are standing on this 'What is a martingale?' stance and the market according to you is a martingale. Then all your research is over, you can't even make any money theoretically. But there are those who do not believe in this and look for patterns and try to mathematically describe them to invest in ATS.
Faith is the last to die.
Quite right, we believe there are patterns. I wonder if the methods proposed here assume anything else. Sort of by default, implicitly. For example, that it is possible to calculate some distribution parameters and reconstruct the hypersurface based on them?
There. These are already the right questions! The answers are unfortunately not obvious, finding them is difficult and requires some knowledge and ability.
Join the discussion, share your knowledge!
I take it you are standing on this 'What is a martingale?' position and the market according to you is a martingale. Then all your research is over, you can't even make money theoretically. But there are those who do not believe in this and look for patterns and try to mathematically describe them to invest in ATS.
Faith is the last to die.
No, not on this one. Regularities do exist, they can be exploited. The first thing that comes to mind is predicting the price. The fact that I think it is futile to predict the price does not indicate martingale, nor does it at all indicate that there is nothing else to predict. Suppose I'm wrong, why don't you say that it is possible to predict the price by some mathematical theory because the price has just such-and-such mathematical properties that the mathematical theory supports?
Once again. I am not against looking for patterns and describing them mathematically. But any mathematical description, before it describes its great properties, warns about the limitations of its usage, assumptions etc. For example, it warns that the distribution must be normal or the process stationary. Similarly, it is the case with regressions, various analyses and dynamical systems theories.
Actually a simple and specific question, how do you fit the market into the assumptions of the Dynamic Systems Theory, got the answer that the market is a dynamic system. I doubt that such an answer is sufficient to make price predictions. What prevents you from linking Dynamical Systems Theory to price with the underlying assumptions of both theory and price itself?
There. These are already the right questions! The answers, unfortunately, are not obvious, finding them is difficult and requires a certain knowledge and ability.
Join the discussion, share your knowledge!
My knowledge does not allow me to relate Dynamical Systems Theory to price with the underlying assumptions, both about the theory itself and about price itself. You see, let's assume, in a redemptive way, that you have wrongly assumed that this theory is suitable for predicting price. I don't think it is at all. I want to uncover that. So I ask why all of a sudden this theory? If the premise is there (apart from similarities in the name and purpose of the task at hand), then I would be wrong. So I'm already in the discussion.
Oh, what persistence. I gather that your, Vita, knowledge of dynamical systems theory is extremely modest. Otherwise you would know that the theory of dynamical systems allows you to express even such complex and inherently chaotic systems as self-organisation.
Well, let us first return to the basics. What is a system in the sense of the aforementioned theory? A system is any object of nature whose state changes in time according to a certain law. If the market is not such a system, then as it has been rightly pointed out - we have nothing to do here. But we are well-proven optimists, aren't we?
A dynamical system is defined as a system whose state is uniquely determined by initial conditions and time. It makes no sense to pull it on the market in this form and no one here, I hope, is doing it.
However, certain classes of dynamic systems are perfectly suited to modelling chaotic systems capable of self-organisation. And if the problem of identification of parameters of suitable dynamical systems can be solved skillfully, they can successfully play the role of a model of the chaotic system under study.
Now imagine that there are methods of transition from the phase space of the initial system to the phase space of auxiliary systems and by analyzing them with the existing methods it is possible to achieve certain results.
The first thing that comes to mind is to predict the price.
No, well let's forecast the weather outside. But how will you then generate trading signals based on these forecasts?
I, for example, see little point in coaching complex mechanisms, including NS, on indicators that are essentially price transformations. What's the point in that? Give to the NS the same data as to be inputted to the indicator and the NS will adapt to the indicator functionality. So why bother it with unnecessary data?
I have not yet seen anything good that would work in the market, from models based on pure statistics. Why do you think there are so many such models: AR, ARM, ARMA, GARCH, EGARCH... the list goes on to several dozen. They simply do not work, although they solve a much simpler task - volatility prediction. And what to say about practically applicable forecasts?