- Joined
- 1/7/25
- Messages
- 2
- Points
- 1
I came from a CS background and did my Masters in Quantitative Finance at Georgia Tech almost a decade ago. I never got a job in the field, instead did regular software jobs (long story) so I don't have a good sense for what the day-to-day might be like. That being said, I do remember ML being given a decent amount of importance in my program. I took 1 program elective that was entirely focused in ML for trading, a core course where ML algorithms were to be applied to common financial datasets in a sort of capstone project, and a general elective that was also ML-focused). With all the buzz surrounding LLMs, I'm wondering if they have any use or even any usefulness in quantitative finance.
My (possibly wrong) take is they probably aren't very useful. I suspect for your trading system, a well-trained neural network (and probably other algorithms) specifically trained on financial data would probably outperform an LLM even with tons of RAG, but Idk. I'm not an expert on LLMs and have never worked on an actual trading system used on production. What are your thoughts?
My (possibly wrong) take is they probably aren't very useful. I suspect for your trading system, a well-trained neural network (and probably other algorithms) specifically trained on financial data would probably outperform an LLM even with tons of RAG, but Idk. I'm not an expert on LLMs and have never worked on an actual trading system used on production. What are your thoughts?
