thanks a lot for commenting and for the Heilnein :) I definitely agree with you in a way, that from the fundamental knowledge point of view, you have to be a jack-of-all-trades. It simply makes you wiser and aware of totally different things in the profession and in life. I tried to make the tests, involving as more…
Hi, thanks for reading! Of course I could do much more and try different options, but this blog post aims first of all to show to the audience alternative possibilities and some primary empirical evidence of their usefulness. This is not a complete guide on how to trade and make a lot of money :)
Agree on the using the “differencing” term, it technically more correct.
What do you mean by TS technics to non-TS data? I see more the opposite situation: using non-TS technics for TS data myself…
Hi Ricardo, I think that for this kind of problem you don’t really need deep learning since it’s a thing that is usually working even on the end device… Something like this https://www.nxp.com/docs/en/application-note/AN3919.pdf should work just fine
I definitely agree with you on both issues. Concerning the ”supervised” thing, I meant, that you (I guess?) try to model physical or other processes in order to explain some particular behavior. I think I didn’t explain myself clearly, that to me this observed behavior is the “Y” in the classical ML paradigm and you aim to find “Xs” that are able to model it.
Hi birinhos, I think you have misunderstood a bit the scope of this work. It’s related not to re-training a machine learning model that predicts something in the future but to calibration of financial models for option pricing to the current market state. There is a huge difference between these two tasks.
Hi, can you explain please more in detail what do you mean by persistence prediction? If results will be persistent over time? It’s different issue, I’ll write about it in next posts, it’s not trivial, inderd