Forums  > Trading  > Applying machine learning to trading  
     
Page 1 of 1
Display using:  

Bytes32


Total Posts: 5
Joined: Aug 2021
 
Posted: 2021-08-03 10:05
This has been tried and tested 1000's of times before - Is it actually practical though?
Some common obstacles:

- price data is non-stationary
- signal to noise ratio is extremely low
- not enough data

I've trained a Transformer model on trying to predict intra-day trends but it really didn't perform that well out of sample.

rod


Total Posts: 435
Joined: Nov 2006
 
Posted: 2021-08-04 10:40
Related

doomanx


Total Posts: 115
Joined: Jul 2018
 
Posted: 2021-08-04 10:57
'Read EL, read EL, he is the master of us all.'

Just read EspressoLover's posts on ML in trading, it's literally the best advice you'll find about it on the internet.

Also, don't get too hung up on the ML part of it. ML adds a little bit extra on top, but you won't find strategies by just slinging ML at raw data.

did you use VWAP or triple-reinforced GAN execution?
Previous Thread :: Next Thread 
Page 1 of 1