The market is a controlled dynamic system. - page 551
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 market analogy with an aeroplane is quite good, but it's exactly the result of moving from a game model to a dynamic system. The "weather" around the plane itself is not the result of natural elements, but of a huge number of market participants. The transition to treating the actions of a large number of people as the action of some natural element is the most conceptual part of modelling the market. It is worth trying to look for conscious approaches to this part without substituting them solely for intuition.
In this approach, you just need a system of verified sensors, and a decision making algorithm based on them to control the flight, more suitable analogy with missile flying to the target not by GPS, but based on only known to it terrain patterns (altitude map, starry sky, sun position, magnetic fields), and interference from the enemy to try to throw it off course or completely disrupt it. As far as I remember from the university program, the Prolog language is used for such purposes, it was used for automatic flight of the Buran, most likely in military rockets too, or something close to it. But of course, the point is not about the language.
In this approach, you just need a system of verified sensors, and a decision making algorithm based on them to control the flight, more suitable analogy with missile flying to the target not by GPS, but based on only known to it terrain patterns (altitude map, starry sky, sun position, magnetic fields), and interference from the enemy to try to throw it off course or completely disrupt it. As far as I remember from the university program, the Prolog language is used for such purposes, it was used for automatic flight of the Buran, most likely in military rockets too, or something close to it. But the point is not about the language, of course.
If in terms of rockets, it is possible to consider that thousands of people simultaneously launch their rockets. These rockets, of course, interact with the atmosphere and with each other in complex ways (not just randomly colliding, but hunting each other down and / or dodging). We are completely unable to accurately describe the impact of all the missiles on ours. So we build some simplified model, where all these rockets are considered as a part of some notional complicated atmosphere or terrain) In terms of game theory, there is a transition from playing with many human opponents to playing with nature. In fact, this transition is interesting, as it is more or less clear what follows.
The pathos of my statement is a bit different) If in terms of rockets, you can think of thousands of people launching their rockets at the same time. These rockets, of course, interact with the atmosphere and with each other in complex ways (not just randomly colliding, but hunting each other down and/ or dodging each other). We are completely unable to accurately describe the impact of all the missiles on ours. So we build some simplified model, where all these rockets are considered as a part of some notional complicated atmosphere or terrain) In terms of game theory, there is a transition from playing with many human opponents to playing with nature. Actually, this transition is interesting, as it is more or less clear what follows.
I use the following approach: we have a theory, simple, well-known, a strategy, also not complicated, we launch a rocket in a medium, preliminarily describing environment (as much as possible simplified but accurate) and setting a program how to act within the strategy, we look how far a rocket flew and where it was shot down, we study the landscape and other external factors, look for the reason why it fell, we add a new behavior pattern, launch it again - is the result better? If so, save it and run it again. In fact it's not as scary as it may seem at first glance, we don't need neural networks here. All possible film scripts and book plots are reduced to 36 dramatic situations, so all human cultural heritage - 36 patterns, differences in scenery, market is much easier =). You can try for a long time to pick up a scientific theory and mathematical-physical justification, or you can experiment and observe and experiment again. The market has 3 states - trend, flat, and uncertainty (which can also be formalized if you want) it's really simple =)
The analogy of the market with an aeroplane is quite good, but it is precisely the result of moving from a game model to a dynamic system. The "weather" around the plane itself is not the result of natural elements, but of a huge number of market participants. The transition to treating the actions of a large number of people as the action of some natural element is the most conceptual part of modelling the market. It is worth trying to look for conscious approaches to this part without substituting them solely for intuition.
Good comparison) with people - with society, with members of a polynomial - with the polynomial itself. And the general condition is that it is impossible to take into account the state of the members)
OK, let them be stochastic. The point is that in a simplified view the system for the trading tool is known and extremely simple - it is a double auction, the price chart is the result of the engine matching according to the double auction principle reflected on the timeline.
The regularities are not in the system, but roughly speaking, in the control signals, so adding the system to the search complicates the task and contradicts the principle of Occam's razor
Uh-huh.
____
In short, if you want to poke around in this topic, be my guest, but the outcome is likely to be like a tractor.
Doesn't contradict it. And it doesn't make it harder. Averages don't affect anything.
Apparently we have very different views on regularities. On their causes and their consequences).
I use the following approach: there is a theory, simple, common, well-known, there is a strategy, also not complicated, we launch a missile in the environment, pre-describing the environment (as simplified as possible but accurate) and setting the program to act within the strategy, we see how far the missile flew and where it hit, we study the landscape and other external factors, look for the reason why it fell, add new behavior pattern, launch again - the result is better? If so, save it and run it again. In fact it's not as scary as it may seem at first glance, we don't need neural networks here. All possible film scripts and book plots are reduced to 36 dramatic situations, so all human cultural heritage - 36 patterns, differences in scenery, market is much easier =). You can try for a long time to pick up a scientific theory and mathematical-physical justification, or you can experiment and observe and experiment again. The market has 3 states - trend, flat, and uncertainty (which can also be formalized if you want) it's really simple =)
The problem is that those who are struggling with your missiles are also not sitting still and are also constantly improving something in their missiles). Actually, game theory is what appeared in its time for such military applications, when the fight is against people and not against nature. If you only react to the situation, there is always the risk of being too late, and if you try to anticipate, the risk of being too hasty. It is exactly the same with flat and trending.
Good comparison) with people - with society, with members of a polynomial - with the polynomial itself. And the general condition is that it is impossible to account for the state of the members)
Well, yes, there is only a general understanding of what the other players want to get in the end game and an assumption of their rationality) In game theory, this is formalised in the form of a matrix of values and equilibrium theory in the game.