AI trading agents face a reliability bar human apps never did
Picture two AI trading agents running the same model and the same strategy. A signal arrives during a fast market, both send an order, and one platform answers in…
Hi HN,I run Quanterlab which is a quantitative research platform (can as well be described as backtesting tool for trading and investing ideas) and it‘s purpose is to provide interpretable results for users' own rule sets, using point-in-time data.I am coming from mechanical engineering domain and I wanted to reflect the concepts of engineering. For example, the “strategies” or more elaborately put “market inefficiency hypothesis” are constructed as block diagrams. And in its own style, QuanterLab is quite a strange thing for a finance software. Instead of looking at candle sticks, this is to construct a logic with these block diagrams. Each one of the boxes have certain contracts with each other’s and it is all about stopping people to make things that does not make sense.Information leakage is something I care a lot and for that purpose everything constructed is placed automatically on a timeline showing you which piece of the block is using which time window which is intended as a visual safety net.And also, the AI is big part of it, which I am personally happy with my decisions on how it is implemented. AI is allowed to use tools in the platform to construct what you describe…
Discussion (0)
No comments yet. Start the conversation!