Machine learning for robots - page 12
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
The market is constantly changing and the bot on one algorithm will fail and all will fly into the pipe.
And the flat can be divided like this. I haven't seen anything better yet...
Another thing is to sit on the tail of market memory on machine learning, there is nothing to think about, just teach the bot to trade on peaks and troughs on the history of prices.
Of course, you need to teach it quickly and qualitatively, maybe you will have to do it often, but it is all solved by primitive automation, especially since I already have it.
All that remains is to check in practice how much a trained robot can trade by inertia and how often it needs to be changed or retrained, and what parts of history to take.
It's like going down a ski track and jumping off a ski jump, accelerate, jump and fly as long as you can, then go back up the hill, which is even easier:)
That's right. in my variant, inertia is 1-3 days with training over the previous few months. and let neuronka decide which sections to take. it is only necessary to scale the flat correctly - not to stretch the price over the whole matrix. No one in their right mind will use absolute values. and the system will distinguish it. But in general, I have not got stable results so far. There is something else... I haven't tried it on big TFs. There you need to be able to wait out a big drawdown. Who will try large TFs - report on the results.