Machine learning in trading: theory, models, practice and algo-trading - page 2713
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
Let us imagine that you are a mantis and cannot distinguish one fictitious event from another. For you, cause and effect are one event and, therefore, occur simultaneously without any temporal and informational connections between them, since there is no time series and the events themselves are fictitious. You collect such fictitious events, wrap them in leaves, admire your work, tell your friends about them. Your events happen outside of time and space in your head. And you seem to be doing pretty well, but all of this bears little resemblance to machine learning.
Does anyone deny the ability to represent information in time series? No, it's just not enough. And I'm not denying time, why would I--
Does anyone deny the possibility of presenting information in time series? No, it's just not enough. And I'm not denying time, why would I...?
What's not enough? You generate your "Events" from a time series, just take pieces of graph or samples, training examples. You do component decomposition, you do time decomposition.
It doesn't matter what way you divide the original time series and what manipulations you do with it.
You can also throw tomatoes at it, blow on it, cluster it, but no events will fly out of it.
Apparently, the "Event" for you is the fact that you can search for some price patterns, a la the first class of techanalysis
What's not enough? You generate your "Events" from a time series, just take pieces of graph or samples, training examples. You do component decomposition, you do time decomposition. That's all you can do.
It doesn't matter at all how you divide the original time series and what manipulations you do with it.
You can also throw tomatoes at it, blow on it, cluster it, but no events will come out of it.
Price reflects the aggregate actions of traders - the flow of water - a river, and events are discrete actions - filling and draining the river, i.e. changing its course.
It is not about rejection of time series - price flow, but about searching by other tools for points influencing the flow.
Apparently, the "Event" for you is the fact that it is possible to find some price patterns, a la first class of techanalysis
Yes, yes - I believe that people/TS influence price, and use technical analysis INCLUDING technical analysis to influence them.
Price reflects the aggregate actions of traders - the flow of water - of a river, and events - discrete actions - filling and draining the river, i.e. changing its course.
It is not about rejection of time series - price flow, but about searching by other tools for points influencing the flow.
more and more superfluous words
Yes, yes - I believe that people/TS influence price, and they use technical analysis INCLUDING technical analysis to influence price.
You need to define what you are working with, it's all generalities.
either you work with price - time series, or with exogenous sources of information.
on the Forex price chart, the event is the appearance of a new tick in general, other events do not occur there.
your indicator constructions in the form of conditions are not events, as well as patterns.
more and more unnecessary words
Do you think it's easy for me to invent abstractions to help your mind see a bigger picture than it's used to?
you need to define what you're working with, it's all generalities.
either you work with price - time series or exogenous sources of information.
on the Forex price chart, the event is the appearance of a new tick in the general case, no other events occur there
your indicator constructions in the form of conditions are not events, as well as patterns.
Price - traces - by which I want to determine the type of beast and its behaviour.
Why make references to time series theory - yes - purely within the theory - price is the object of study.
But time series are used, for example, to control the work of mechanisms, and there we get information about the mechanism, there are many such mechanisms, whether the conveyor has broken down completely or not can be found out by analysing all these time series.
You see, data from different mechanisms are merged into the glass at once, because they all push the price, but the reason is different for each of them.
Price - tracks - by which I want to determine the type of beast and its behaviour.
Why make references to time series theory - yes - purely within the theory - price is the object of study.
However, time series are used, for example, to control the work of mechanisms, and there we get information about the mechanism, there are many such mechanisms, whether the conveyor has broken down completely or not you can find out by analysing all these time series.
You see, data from different mechanisms are merged into the glass at once, because they all push the price, but the reason is different for each of them.
because it's called time series classification, there's no other term for it.
how it's done is irrelevant. You collect data, you do any manipulation (which are not events) in order to make a prediction.
it's all data, not events.
And you think it's easy for me to make up abstractions to help your mind see a bigger picture than it's used to?