@freemo Great. So is it a learning-based algorithm like RL where you model each timestep as an immediate reward and use a model-free algorithm to keep it generic to any market and optimised with specific short-term discount factor for daily profits?
Apologies if the question is too basic since I'm new to algorithmic trading.
@gokulnc Sadly I cant offer up much detail on it. It makes me a lot of money and if i give away how it works I could loose that advantage should it get out to the general public.
Best I can do is tell you i can point it any stock and it tells me when to buy and sell and consistently increases my profits over buy and hold
@freemo Thanks, I understand. It would be interesting to know if you include any other context into your algorithm in addition to the history of prices in the market.
If not, it might mean you're good at picking the right stock to invest in (which magically goes well with the algorithm you have which optimises for short term cycles of buy-and-sell for some short amount of time/days) 🤔
@gokulnc The benchmarks I do on my algo is on the TQQQ v. SQQQ swing specifically so i can rule out any influence from my stock selection bias.
@matrix Thats a good strategy to use. I do my own personal stock picks and I use an algorithm I invented. Right now i kinda use them together. I pick stocks I think is good then use my algorithm to tell me if and when im alowed to hold it or not to ensure i only hold it on up turns