Machine learning in trading: theory, models, practice and algo-trading - page 1834

 
mytarmailS:

For example, we cluster the price into 10 clusters...

What is a cluster - it is a kind of image that repeats - a "pattern"

The cluster has a number v. dn. case from 1 to 10

The price in the form of clusters will look like this

1113333555433377779991010103333222288888

we choose the combination at random, for example 1593

look for it in the price

1113333555433377779991010103333222288888

this will be the resultant pattern

put all situations with this pattern together and look at the statistics of profitability


there may be hundreds of clusters, you can build them on any data, you can replace clusters with log rules for example, it will be more clear

At random it is overkill (random or otherwise is another question) and there are a lot of variations even in 10 values. And they are not necessarily profitable if random choice. Profitability logic is needed, it will reduce the number of choices.

 
Maxim Dmitrievsky:

And if there is a strong trend, you will have almost no prices in the range

and always a different number of features turns out

Accordingly, one algorithm / model can not do, you need a criterion for selecting models. Maybe it depends on the number of features)))

 
Maxim Dmitrievsky:

And if you have a strong trend, you'll have almost no prices in the range

and always a different number of features

here's a strong trend for you.

What's changed?

 
mytarmailS:

For example, we cluster the price into 10 clusters...

What is a cluster - it is a kind of image that repeats - "pattern"

Here you can be more detailed, clustering the price on 10 clusters what does it mean? Because I'm confused. To me it means to divide the price into 10 parts, but apparently not so, since a cluster is an image.

 
Valeriy Yastremskiy:

Here you can be more detailed, clustering the price by 10 clusters what does it mean? Because I'm confused. To me it is to divide the price into 10 parts, but apparently not so, since the cluster is an image.

Are you familiar with clustering?

 
Valeriy Yastremskiy:

Overkill is overkill (random or otherwise is another question).

Overkill is when you have more than 21 where Jack-2 queens-3 kings-4. Everything else is evil :-)

 
mytarmailS:

Are you familiar with clustering?

in the sense that it is the division of a set into subsets. As here apparently not.

in the idea that the pink area should be divided into 10 parts apparently)

 
You're all such a drag lately. It's boring with you. Trickster, where are you... let's talk....
 
Valeriy Yastremskiy:

in the understanding that it is a division of a set into subsets. As here apparently not.

All true is the division of a set into a specific number of classes. That is, the classes must be defined or not. Say you have 100 strings, but you can group them into 10 groups... as an example.

Classification, on the other hand, only affects 3 classes at most. "Yes", "No", "Don't know".

 
Mihail Marchukajtes:
What a drag you all have been on lately. Bored with you. Trickster, where are you... let's have a chat....

Yes, the grail theme there seems to be not quieted down, there are crackling people there))))