Machine learning in trading: theory, models, practice and algo-trading - page 2716

New comment
 
Valeriy Yastremskiy #:

The result of data processing is the same data. The dimensionality and type of data are the same. It is possible to assume that behind some sequences of prices there are some events and call this sequence an event, but in my opinion the term will be confusing. This is too general understanding of events, in the common understanding it is still concrete actions). And the type of these events and prices are quite different.

I am not imposing anything.

Earlier I wrote and showed that Events activate actions. Events themselves are a consequence of other events.

 
Aleksey Vyazmikin #:

I'm not imposing anything.

I wrote and showed earlier that Events activate actions. Events themselves are a consequence of other events.

Then a certain result of the totality of events))))

 
Valeriy Yastremskiy #:

Then a definite result of a set of events))))

The events will be fed into the model for training, and there they will be combined :)

 
Aleksey Vyazmikin #:

Events will be fed into the model for training, and there they will already copulate :)

So come on, feed it in and show me what you got, how many times can we talk about nothing?

 
mytarmailS #:

Well, come on and show us what you've got, we can't go on and on about nothing.

You have to do it that way - and discuss the details of realisation, and there's no one to discuss it with.

So I'm leaving. I've wasted too much time on sermons.

 
Valeriy Yastremskiy #:

The result of data processing is the same data. The dimensionality and type of data are the same. It is possible to assume that behind some sequences of prices there are some events and call this sequence an event, but in my opinion the term will be confusing. This is too general understanding of events, in the common understanding it is still concrete actions). And the type of these events and prices are quite different.

For some reason nobody calls birch bark an event. Birch bark is an object with properties. It can be divided into elements. It belongs to the class "tree bark". The connection of parts of the bark to each other is not an event. If something in the bark explains something else in the bark, it is also not an event.

A price graph is also an object with properties, it belongs to the class "time series". Parts of the chart in any of their transformations are not events. Relationships between parts of the chart are not events either. If some part of the chart explains another part, it is not an event.

Classification of a time series is the division of parts of the graph by their properties.

The formulation that a condition on the chart or the value of an indicator reflects other events is untenable in a practical sense, because it is not signed what events it reflects at a particular moment in time. These can be events of different types, overlapping of events, and the condition and reaction in the form of rate change can be the same for all cases.

Again there will be substitution of notions and violation of logical connections, as if one type of events led to the same consequences, although the types of events were different. On the contrary, identical events can lead to different consequences.

And if the opposite is asserted, the events should be signed and added to the model as exogenous.

He also voiced the option that condition + consequence are one event. This is a violation of the law of identity.

That is, no logic is traced at any stage at all, but how much self-confidence.
 
Maxim Dmitrievsky #:
For some reason nobody calls birch bark an event. Birch bark is an object with properties. It can be divided into elements. It belongs to the class "tree bark". The connection of parts of the bark to each other is not an event. If something in the bark explains something else in the bark, it is also not an event.

A price graph is also an object with properties, it belongs to the class "time series". Parts of the chart in any of their transformations are not events. Relationships between parts of the chart are not events either. If some part of the chart explains another part, it is not an event.

Classification of time series is a division of parts of the graph by their properties.

The formulation that a condition on the chart or the value of an indicator reflects other events is untenable in a practical sense, because it is not signed which events it reflects at a particular moment in time. These can be events of different types, overlapping of events, and the condition and reaction in the form of rate change can be the same for all cases.

Again there will be substitution of notions and violation of logical connections, as if one type of events led to the same consequences, although the types of events were different. On the contrary, the same events can lead to different consequences.

And if the opposite is asserted, the events should be signed and added to the model as exogenous.

He also voiced the option that condition + consequence are one event. This is a violation of the law of identity.

That is, no logic is traced at any stage at all, but how much self-confidence.

Well, at least it became clear what is understood by the events in Alexei. In general, there is nothing very critical in the difference of terms, the main thing is to understand them, of course it is better to understand them to the end, but in this environment it is difficult. There is no guarantee that other terms are interpreted in the same way.

In general, an event is a price pattern, found as a signal and as an apparent trend or maybe a wave. And so I understand that these found patterns go into the next cycle of training as reference ones, if I understand everything correctly, and then a new set of pattern events is made, with the hope that these new rules will be better than the old ones in terms of profit.

 
Valeriy Yastremskiy #:

Well, at least it became clear what is understood by the events in Alexei. In general, there is nothing very critical in the difference of terms, the main thing is to understand them, of course it is better to understand them to the end, but in this environment it is difficult. There is no guarantee that the other terms are interpreted in the same way.

In general, an event is a price pattern, found as a signal and as an apparent trend or maybe a wave. And so I understand that these found patterns go into the next cycle of training as reference ones, if of course I understand everything correctly, and then a new set of pattern events is made, with the hope that these new rules will be better than the old ones in terms of profit.

It's a mess again. Rubbish in, rubbish out.

All one had to do was to define precisely the object being studied. Then define the method.

I am amazed to see how technicians by education get confused in their own thoughts.
 
Maxim Dmitrievsky #:
For some reason nobody calls birch bark an event. Birch bark is an object with properties.
The bark of a birch tree is a consequence of the event "someone planted this birch tree many years ago".
It's the same thing on a price chart.
 

I don't think categorical judgement is good for communication on the topic. In depth, we often do not know what is primary, whether an egg is a cause or a chicken is a consequence. Events do not describe an object, properties of an object are often consequences of some causes - actions - events.

But, the name of an object does not usually carry its properties. Well, unless the names are not ranked by properties. After all, the object is described by its properties. A price series is the only object that can be primary described and fixed in the form of ticks. Predictors, clusters, quanta, etc. are terms inside some algorithms of processing this data.

In general, the data is only tick prices, time or tick sequence number, and further filters or averaging are data. The result of processing these data are the same prices, time or serial number of a tick or bar, or their sequence.

Well, if it is a significant sequence, then we can call it an event))))

Although it reminds me of the story of the machete. When I saw it, I did not know that it was a machete, and I called it a long axe and for several years I was sure that it was a long axe))) until I learnt that it is a machete in other countries)))))

1...270927102711271227132714271527162717271827192720272127222723...3656
New comment