From theory to practice - page 702
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
The use of non-stationary (but piecewise stationary) SB is sensible enough. It is suitable for trends and their changes. For prices in a corridor, for example, something else is needed (e.g. stationary with dependence and non-zero ACF). So yes, it is unlikely that a theorver can give some kind of Uniform Price Model.
But on the other hand, we have no other meaningful ways of dealing with uncertainty.
Wrong, there are"meaningful ways of dealing with uncertain ty".
But you're trapped in the framework of TViMS, and you can't get out of that corridor, you're trapped. And this prevents you from seeing how diverse the world outside your corridor is.
Wrong, there are meaningful ways of dealing with uncertainty.
But you're trapped in the framework of TViMS, and you can't get out of that corridor, you're trapped in it. And this prevents you from seeing how diverse the world outside your corridor is.
Oleg, why not, theorist and matstatistics can handle it, and as for how diverse the world is, interestingly, I would also like to know, it is necessary to develop))
What does "it" cope with?
what does "it" deal with?
Meaningful ways of dealing with uncertainty, so in the text.
Forum on trading, automated trading systems and testing trading strategies
From Theory to Practice
Aleksey Nikolayev, 2018.10.31 16:08
The application of non-stationary (but piecewise-stationary) SB is quite meaningful. It is suitable for trends and shifts. For prices in a corridor, for example, you need something else already (e.g. stationary with dependence and non-zero ACF). So yes, it is unlikely that a theorver can give some kind of Uniform Price Model.
But on the other hand, we have no other meaningful way of dealing with uncertainty.
Forum on trading, automated trading systems and testing of trading strategies
From theory to practice
Oleg avtomat, 2018.10.31 16:58
Wrong, there are such meaningful ways of dealing with uncertainty.
But you are constrained by the framework of TViMS, and you can't get out of that corridor, you are trapped. And it prevents you from seeing how diverse the world outside your corridor is.
have you lost the thread?
Wrong, there are meaningful ways of dealing with uncertainty.
But you're trapped in the framework of TViMS, and you can't get out of that corridor, you're trapped in it. And that prevents you from seeing how diverse the world outside your corridor is.
Philosophically, you're right. The randomness studied by theorist is a very special case of the general concept of uncertainty. For example, very different kinds of uncertainty are dealt with in game theory or dynamical systems theory. But as soon as it comes to solving meaningful problems, many basic methods in these areas turn out to be probabilistic in nature. These are Nash equilibria in TC or stochastic DM in DC.
have you lost the thread?
Philosophically, you are right. The randomness studied by a theorist is a very special case of the general concept of uncertainty. For example, very different kinds of uncertainty are dealt with in game theory or dynamical systems theory. But as soon as it comes to solving meaningful problems, many basic methods in these areas turn out to be probabilistic in nature. These are Nash equilibria in TI or stochastic DMs in DS.
No, not really. Not in their nature, but in their description, so that you can have something to fall back on. And that's not the same thing at all. Although, if you don't get into it, it seems to be exactly as you say.
For example, when creating an adaptive system, I need to take into account the influence of interference, whose behaviour is unknown and could be anything (within max and min tolerances), and what its distribution will be in the future is unknown anyway. When building a system, I accept(assign) any interference distribution that is convenient for me. Convenient in terms of interference compensation by the system automatically. This is a mathematical trick. Ultimately, the adaptive system built works in the presence of interference with any distributions, not only with those adopted at the stage of formalizing the problem. And in this case there is no identification of the interference distribution, as there is no need in that.
But using methods of the theory of adaptive systems this problem is quite solvable and allows for further processing.
Well, stochastic DM in DC is just one of the sections of the theory that gives this tool in its hands, among others.
No, not really. Not by nature, but by description, so that you have something to fall back on. And that is not the same thing at all. Although, if you don't get into it, it seems to be exactly as you say.
For example, when creating an adaptive system, I need to take into account the influence of interference, whose behaviour is unknown and could be anything (within max and min tolerances), and what its distribution will be in the future is unknown anyway. When building a system, I accept(assign) any interference distribution that is convenient for me. Convenient in terms of interference compensation by the system automatically. This is a mathematical trick. Ultimately, the adaptive system built works in the presence of interference with any distributions, not only with those adopted at the stage of formalizing the problem. And there is no identification of the interference distribution, as it is not needed.
Nevertheless, there is no way to construct a stochastic DM apparatus (starting from Ito and Stratonovich integrals) outside the framework of probability theory. What you are talking about is the subtleties of applying the apparatus, not creating it.