Article: Price forecasting with neural networks - page 7
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
OK, no more arguments about molecules and statistical physics. After all, it is off-topic for this forum and even this thread... I wish you success in inflating balloons with one molecule and calculating temperature and pressure of one molecule.
P.S. I'm not a mathematician, that's just my nickname...
The article is, of course, a little populist and bravura, but the pictures are nice and some of the links are interesting. I, for instance, found out that it's easy to export data and integrate applications into Matlab. - But I really didn't know about it. In general I think the authors of the article should get a grant from Matlab for promoting and popularizing the product.
The article contains some very prominent pearls like - "It is known that 99% of all transactions - speculative, ie, not aimed at the service of real turnover, but designed to make a profit on the scheme "bought cheaper - sold more expensive. Well, yes it seems to be on the subject of the article is not very relevant to the subject and probably because it passed.
Now actually on the subject - ANN is a tool that can be used to find correlations between input signals and output states (actions). Importantly, an ANN has a very large number of adjustable parameters - weights
synapses. As a consequence the Achilles' heel of ANN is overoptimization. Actually this is what the authors of the article write about - at a training interval they have a fairly high depth of successful prediction, and beyond this interval it quickly goes down. But nevertheless in spite of a strong "tendency" to overoptimization ANN is a very interesting tool for finding
"fuzzy, fuzzy patterns, if you know how to prepare it (ANN).
The hardest part, in my opinion, is preparing the input data and understanding exactly what kind of patterns the network should find, as well as how to avoid "dumping" it into overoptimization.
In my opinion, using the standard means of MetaTrader we can also create ANN - by integrating it in indicators. In the loop it starts the learning process at a certain interval, then the forecast is drawn, then the interval is shifted and retraining takes place again, etc. This will result in a kind of intelligent, self-learning indicator. I haven't really done it, primarily because of the complexity of preparing the input data and problems with over-optimization.
As a consequence, the Achilles' heel of ANN is over-optimisation.
For example Classification or Pattern Recognition. A properly formulated problem is solved perfectly with these methods.
The mathematician is right - the concepts of volume, pressure and temperature simply do not exist for a single molecule.
And since they don't exist, there's nothing to talk about.
"If the parts in principle do not possess some property, even in embryo, then the system of those parts will not possess that property either. " is not true.
Ultimately everything is made up of elementary particles, which are made up of quarks, and at the level of quarks there are no concepts of the macrocosm except the laws of conservation of energy, momentum, momentum ... And even then, at the level of elementary particles, even these fundamental laws are approximately satisfied (because of the uncertainty principle). There are plenty of things that arise only from collective interaction, and are completely absent from single instances. There is even a science called synergetics that studies all this.
"Can anyone here give, give a clear definition of what intelligence is?" - a definition may be given (as many people as there are definitions :)). But what is the point? Till now nobody could determine what is the difference between alive and inanimate? What is the difference between reasonable and unreasonable?
The name Artificial Intelligence was created rather emotionally.
To define it, one must first define natural intelligence,
and a definition of the differences - how artificial differs from natural.
Unfortunately, the existence of God has not yet been proven.
It rejects atheism as unfounded. And it points out that there is something else, and that's about it :).
Atheism is grounded in a scientific approach to comprehending reality.
One of the fundamental theses of the scientific approach: "The world is conceivable". And it is indeed conceivable! The proof of it is successes of technological civilization in expansion of material possibilities of a man.
However, the opposite is also true: "The world is incomprehensible". If only because in order to comprehend the world one needs infinite time but neither an individual nor the whole mankind can afford it. But, of course, this is not the only argument and not the strongest one.
Here you have dialectics, the high philosophy, which so many people disparage. As a result, they inevitably end up in a blind alley of the limitations of the scientific approach, in a delusion about the power of rational knowledge. Or simply in the captivity of the primitive "what I don't see isn't there". :-)
Break guys :))
The mathematician is right - the concepts of volume, pressure and temperature simply do not exist for a single molecule.
And since they don't exist, there's nothing to talk about.
"If the parts in principle do not possess some property, even in embryo, then the system of those parts will not possess that property either. " is not true.
Ultimately everything is made up of elementary particles, which are made up of quarks, and at the level of quarks there are no concepts of the macrocosm except the laws of conservation of energy, momentum, momentum ... And even then, at the level of elementary particles, even these fundamental laws are approximately satisfied (because of the uncertainty principle). There are plenty of things that only arise from collective interaction, and are completely absent from single instances. There is even a science called synergetics that studies all this.
"Can anyone here give, give a clear definition of what intelligence is?" - a definition may be given (as many people as there are definitions :)). But what is the point? Till now nobody could determine what is the difference between alive and inanimate? What is the difference between reasonable and unreasonable?
The name Artificial Intelligence was created rather emotionally.
To define it, one must first define natural intelligence,
and a definition of the differences - how artificial differs from natural.
So, how is it not known?
Living things have perception. Sleep, for example.
===================================================================
But in "collective" interaction, every quark behaves like a quark and there is no other reality for it.
Where does the super quark come from?
As a consequence, the Achilles' heel of ANN is over-optimisation.
- But in this particular thread and in the particular article under discussion, that's exactly what it's about.
Or am I wrong?
A properly formulated problem is solved perfectly with these methods.
- I'm not going to argue against it - the problem is how to correctly
The problem is how to formulate the task correctly or how to prepare data for the grid.
Unfortunately, the existence of God has not yet been proven.
It rejects atheism as unfounded. And it points out that there is something else, and that's about it :).
Atheism is grounded in a scientific approach to comprehending reality.
One of the fundamental theses of the scientific approach: "The world is conceivable". And it is indeed conceivable! The proof of it is successes of technological civilization in expansion of material possibilities of a man.
However, the opposite is also true: "The world is incomprehensible". If only because in order to comprehend the world one needs infinite time but neither an individual nor the whole mankind has this time. But, of course, this is not the only argument and not the strongest one.
Here you have dialectics, the high philosophy, which so many people disparage. As a result, they inevitably end up in a blind alley of the limitations of the scientific approach, in a delusion about the power of rational knowledge. Or simply in the captivity of the primitive "what I don't see isn't there". :-)
It would be so if everything were scientifically, algorithmically, by laws.
But if there is something that cannot be described by the laws of nature at all, then atheism loses its basis.
And such a thing, exists. The existence of perception.
Any biological system, as a mechanical system, should not have perception. Although, to an external observer, it can communicate.