-
-
Notifications
You must be signed in to change notification settings - Fork 189
Description
Hello everyone, I'm just trying to get a foothold in your wonderful project.
My actual goal is to create something similar to this one in cassandre (https://de.tradingview.com/script/SfyBCHIJ/).
I have now discovered that ta4j is no longer included, which is why I am implementing the technical indicators myself in order to use their results as a feature with the help of the Lorentzian distance in the knn near neighbors for buying and selling decisions.
Per se I'm on the right track (I think), but I'm having a hard time with backtesting.
I have to test very large amounts of possible variables (both those of the technical indicators as well as the knn distances and neighboring quantities) - so I'm currently trying to parallelize the backtesting in some way in a parameterized manner.
However, I can't come up with a practical approach.
I would like to ask questions about this in the Discord, but I noticed that it has been switched off.
That's why I would like to ask how you would approach such a problem or perhaps even solved it.
If there is still an active community for exchange somewhere, I would be grateful for that too.
Thank you very much and a Merry Christmas,
Alfons