I may say this presumptiously but I always laugh when people laud AI and ML in trading. The whole key is picking your features, no matter how fancy your model is you put in garbage, you get out garbage.
I think this is true for like 99% of people using ML in finance (including me) but I know algo at least two places using non-engineered features in an ML pipeline and having enormous success with it.
-
-
Don't give away my astrology like that
-
What are your signs?
#astrologyformen
End of conversation
New conversation -
-
-
I'd say this applies to all ml really, not just finance
-
Doesn’t seem to be true for state of the art in applications like image recognition (input is raw pixels) or generative text (input is characters) or reinforcement learning for playing games (input is raw encoding of game state)
- Show replies
New conversation -
-
-
ML seems easier to use in finance for non-trading functions. I've seen some examples of using RNNs for risk calculations
-
Would be interested to know if they give better risk forecasts than like... a low-dim covariance matrix fitted to historical data with some fatness added to the tails
- Show replies
New conversation -
-
-
Quite a few I know dusting off their failed "NN kitchen sink data mining projects" and turning them to Crypto with moderate early success Not sure I rate their chances very high long term, but it's a thing
-
Less established markets are well known for price inefficiency
- Show replies
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.