[alexmingoia]

The problem I see with ML on stock prices is that stock price is not a function of the various available numerical indicators. Markets are non-deterministic and chaotic systems. Contrast that with driving, where the decision to turn left, right, throttle, etc. is largely a function of the features (surroundings, current speed, etc.).

That said, people are using ML to predict future prices, although no hedge fund claiming this has published their returns.

With regards to reinforcement learning, why not just try it? Numer.ai is an ML competition that gives you free obfuscated stock data and rewards you if your model is successful.

[critiq]

Yeah thanks, I have tried, I was comparing and found reinforcement learning is not really reinforcement in this setting. As it cannot allow exploring like what if scenario. It is similar to supervised learning setting as we do in labeling large data and trying to classify etc.

In contrast if we take https://gym.openai.com/envs/Pendulum-v0/ Environment allows algorithm to play with it and learn from new conditions and actions, so is reinforcement. To me finance is like teacher trying to tell same story no matter what student does/learns, bit wierd.