Machine learning in trading: theory, models, practice and algo-trading - page 3544

New comment
 
Maxim Dmitrievsky #:
I came up with this sequence.
1. predict 1 bar ahead, see the error on the new data. So for all sets of signs.
2. choose the model with the smaller error.
3. Do clustering/filtering of predictions, pick the cluster with the lowest error where the model is least wrong.
4. Do sub-clustering on the clusters already found, if you didn't find anything in step 3.
5. You come up with trading rules for the best cluster. At the same time, you can shift the model back 1 bar, if it is easier for you to get signals.

Since the model is linear and therefore fast, you can loop through the fixtures and their combinations or optimisation.

This is roughly the approach I use for classification. It is logical and nechural and understandable.

You end up sifting through a sieve of all the randomness and leaving the signals. There may be variations, this is just a general outline.
1. Why predict? It's pure approximation or interpolation, the model doesn't look into the future.


I was thinking of just forcing Forrest and selecting the best rules, but I realise that the target is not so clear-cut.
 
mytarmailS #:
AMO only has access to one last fly and last RSI value... ALL!!!!

Then how do you get an OHLC bar? Do you forecast like ticks or smaller TF and then roll it into a bar?

 
Aleksey Vyazmikin #:

Then how do you get an OHLC bar? Do you forecast like ticks or smaller TF and then roll up into a bar?

No.

It's just that the model has four outputs, ohlc, respectively.


You didn't even know there was such a thing, did you? :)

 
mytarmailS #:
1. And why predict, it is pure approximation or interpolation, the model does not look into the future.


I was thinking of just forcing Forrest and selecting the best rules, but I realise that the target is not so clear-cut.
Maybe there's no point in predicting. I'm just sketching out some options that can be tested.
 
mytarmailS #:

no.

It's just that the model has 4 outputs, ohlc respectively.

If it's not a wooden model, I can't understand how HL is calculated. From two values you get four.

 
mytarmailS #:
You didn't even know it was like that, did you? :)

There's a lot I don't know.

 
Aleksey Vyazmikin #:

If it is not a wooden model, I cannot understand how the HL is calculated. You get four values from two values.

Two traits

4 targets

4 predictions

Aleksey Vyazmikin #:

I don't know a lot of things.

We're all in this boat.

 
Maxim Dmitrievsky #:
Maybe there's no point in predicting. I'm just sketching out some options that could be tested.

realised

 
mytarmailS #:

Two signs

4 targets

4 predictions

We're all in this boat

Can you show me the content of the model?

 
Aleksey Vyazmikin #:

Can you show the content of the model?

what exactly are you interested in?

1...353735383539354035413542354335443545354635473548354935503551...3634
New comment