The market is a controlled dynamic system. - page 346
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
In game theory, as far as I remember, the economy is treated as an auction, with all that implies. No other models are envisaged there. Arbitrage (statistical) as a temporary inefficiency and path to market efficiency is the most common strategy formulated.
The models cannot be the same for different markets - commodity and speculative. The video is a cooperative game (or its solution), it seems, and we have it rather antagonistic. In economic theory, there are all sorts of models for all occasions, and new ones are constantly appearing. Why don't we come up with a few?)
I don't really understand your terminology. For example, is a stochastic game(Nash equilibrium is there) deterministic to you or not? If all moves in a game are non-deterministic (completely random), then that game naturally does not belong to the realm of game theory. But in the marketplace, it seems to me, not all moves are random.
Again, even in the case of nature games there is no "just optimisation" - a change in the optimisation criterion can dramatically change the optimal strategy and there is no way to uniquely select this criterion. In a game with humans, things are much more complicated.
Game theory allows building models of uncertainty arising not due to blind chance (as in the theorist), but due to opposition of other people. Here is an article that compares these types of uncertainty. The author must have a good understanding of games and markets, by the looks of it.
Well stochastic, who cares. The equilibrium will only work on history, as usual. On the real, nothing will work unless real patterns such as arbitrage patterns are exploited.
This is the very basis for learning with reinforcement, game theory.Read game theory carefully. It only works when the opponent's strategy is unchanged (the rules of the game all the more so), and for multiple players there is often no solution at all.
You are talking nonsense.
There is always a solution - read up to Nash equilibrium. Of course, mixed strategies don't always make sense, but in our case it can be found.
In addition, the TS is about the game with nature (environment) - there, on the contrary, there is a problem with choosing from the abundance of solutions (what exactly the ratio of risk and winnings to optimize). In fact, the optimization of the Expert Advisor is an example of a problem from this area.
Quite right. There is an abundance of solutions. And consequently, there is a problem of choice.
Well stochastic, who cares. The equilibrium will only work on history, as usual. Nothing will work on the real unless real patterns like arbitrage patterns are exploited.
This is the very basis for learning with reinforcement, game theory this.The point is not to look for an equilibrium. For example, we can look at stochastic games as hidden Markov models. There's no getting away from learning from history, but it's better to train more or less interpretable models than poorly interpretable ones (e.g. neural networks)
The point is not to look for equilibrium. For example, one could consider stochastic games as hidden Markov models. There is no getting away from learning from history, but it is better to teach more or less interpretable models than poorly interpretable ones (e.g. neural networks).
You are listening but not hearing. It doesn't work in the market, like everything else in this thread :)
Just saving time.
Forex is a game with a tiger's tail. If you take the market as a game, you are bound to encounter the face of the tiger. No one will seriously play such a high-risk game with the market. You throw that out of your head. The market is governed by economic laws. In a normal market, in order to adequately describe the profit formula, you have to account for 17 real and virtual varieties of price, estimate, using actual data, the maximum virtual volume of the market and a coefficient that takes supply and demand for a given product into account. If one of them is not taken into account, you will never achieve equality of profit according to this formula and the obvious formula as the difference between income and all kinds of fixed and variable costs. In financial markets the task of determining these market parameters is complicated, but, indirectly, they can be estimated and have a positive effect.
I have suggested to Oleg many times that the idea of market management should be abandoned. Management and the market are two mutually exclusive concepts, they are antipodes. If the market can be managed, it is no longer a market.
Yusuf, I've told you many times that it's not about me trying to manage the market. ;))) I am talking about the fact that the market lives and develops under a generalised controlling force, the elements of which may be regulations, currency interventions, news, ratings, etc., etc.
A market without governance cannot exist.
It is unacceptable for you to have some kind of external governance in relation to the market. OK, let's say that's true, but then we have to accept that the market is a self-organising system. And any self-organising system necessarily has a feedback loop, there must be some criterion for its functioning -- and this is control in general terms.
Nash equilibrium is only for deterministic games with known outcomes. Everything else is ordinary optimization and how (and why) to involve game theory, I do not know.
Exactly, you don't.
and why formulate anything when everything was formulated a long time ago by several generations of traders. There's like nothing else to come up with, except your own new derivatives
another nonsense
Name the main views on the market formulated long ago by several generations of traders. They, by and large, do not exist, which is why all the woes of traders.
That's right. No.