As the saying goes, those who know don't say and those who say don't know.
Multi-Agents LLM Financial Trading Framework
31–40 of 90 posts
Re: Multi-Agents LLM Financial Trading Framework
#32Earlier quoted context omitted.
I think this might work. When I started working no the trading agent I mentioned above I wanted to see if it can be just a better investor over the long run. The intention was not to do high-frequency trading. As you can see most of the days it is not taking any actions. The losses where down to mistakenly setting the stop losses too close to the top. If it wasn't so careful it might have made some money tbf. My gut…
Lmao, you llm people have some crazy delusions. You realize markets are zero sum, and if you're using a public model that everyone else also has access too, you llm psychos will destory eachothers "agentic" edge (not that there ever was one). Not to mention all the other obvious flaws with llms, lime having an effective memory of ~200k words and no ability to judge whats actually going on in the real world.
You really have no idea what you are talking about.
I have been running an intermittent experiment with a multi agent "investment firm" for over a year now across model releases.
They certainly can beat indexes, BUT.. the model families have some biases that you have to design around. The stop loss that bit the parent is certainly one. The models like to create rules. Often rules, one of those is making all kinds of exit conditions.
Another big one from my experience is the bias to inaction in a scenario with risk. This means a model without structure around it will bias to keeping too much cash.
Re: Multi-Agents LLM Financial Trading Framework
#33Earlier quoted context omitted.
Lmao, you llm people have some crazy delusions. You realize markets are zero sum, and if you're using a public model that everyone else also has access too, you llm psychos will destory eachothers "agentic" edge (not that there ever was one). Not to mention all the other obvious flaws with llms, lime having an effective memory of ~200k words and no ability to judge whats actually going on in the real world.
You really have no idea what you are talking about. I have been running an intermittent experiment with a multi agent "investment firm" for over a year now across model releases. They certainly can beat indexes, BUT.. the model families have some biases that you have to design around. The stop loss that bit the parent is certainly one. The models like to create rules. Often rules, one of those is making all kinds of…
This would be nice if it was supported by some hard data. But even if it was, one year is too short of a timeline to make any kind of reasonable conclusions about its efficacy.
Buying stock based on coin flips can beat indexes short term too, that does not mean it is a better strategy or that it works over the long term.
Re: Multi-Agents LLM Financial Trading Framework
#34Earlier quoted context omitted.
I think this might work. When I started working no the trading agent I mentioned above I wanted to see if it can be just a better investor over the long run. The intention was not to do high-frequency trading. As you can see most of the days it is not taking any actions. The losses where down to mistakenly setting the stop losses too close to the top. If it wasn't so careful it might have made some money tbf. My gut…
Lmao, you llm people have some crazy delusions. You realize markets are zero sum, and if you're using a public model that everyone else also has access too, you llm psychos will destory eachothers "agentic" edge (not that there ever was one). Not to mention all the other obvious flaws with llms, lime having an effective memory of ~200k words and no ability to judge whats actually going on in the real world.
The 'edge' is holding the investments over long periods of times. Agents are merely automating the portfolio managing part for lower costs.
Re: Multi-Agents LLM Financial Trading Framework
#35Having worked in hedge funds for the last decade, this seems to miss the mark. Firstly we often reward skillstacking ie a technical person later becoming a trader. The more one person knows the better. These people are rare though hence the reason there is still many seperate job functions, so a person can specialize. But an AI agent? They all have the same brain, so why nerf them by specialising. Secondly, browsing…
Actually there is a really good reason, it has to do generally with making sure that all the aspects that you want evaluated are actually evaluated.
Now it may be possible with models like Astra that you no longer need to do this, but in earlier models it was beneficial.
So I might want a macro economic read which leads to a market thesis. Then I would hunt for exposure, then evaluate the candidates across different aspects. Breaking the process up at least made sure no steps were missed and the different aspects considered.
Re: Multi-Agents LLM Financial Trading Framework
#36[flagged]
Re: Multi-Agents LLM Financial Trading Framework
#37[flagged]
Re: Multi-Agents LLM Financial Trading Framework
#38Earlier quoted context omitted.
Lmao, you llm people have some crazy delusions. You realize markets are zero sum, and if you're using a public model that everyone else also has access too, you llm psychos will destory eachothers "agentic" edge (not that there ever was one). Not to mention all the other obvious flaws with llms, lime having an effective memory of ~200k words and no ability to judge whats actually going on in the real world.
You really have no idea what you are talking about. I have been running an intermittent experiment with a multi agent "investment firm" for over a year now across model releases. They certainly can beat indexes, BUT.. the model families have some biases that you have to design around. The stop loss that bit the parent is certainly one. The models like to create rules. Often rules, one of those is making all kinds of…
Come to me when you have 500 trades and can beat Vangaurd's-VOO over a multi year time frame. Ill bet my entire networth and all future earnings for the rest of my life that your bot doesnt beat it. Your llm induced Dunning Kruger is going to get you in trouble one of these days I promise.
Re: Multi-Agents LLM Financial Trading Framework
#39Earlier quoted context omitted.
Lmao, you llm people have some crazy delusions. You realize markets are zero sum, and if you're using a public model that everyone else also has access too, you llm psychos will destory eachothers "agentic" edge (not that there ever was one). Not to mention all the other obvious flaws with llms, lime having an effective memory of ~200k words and no ability to judge whats actually going on in the real world.
You really have no idea what you are talking about. I have been running an intermittent experiment with a multi agent "investment firm" for over a year now across model releases. They certainly can beat indexes, BUT.. the model families have some biases that you have to design around. The stop loss that bit the parent is certainly one. The models like to create rules. Often rules, one of those is making all kinds of…
Is it opensource?
Re: Multi-Agents LLM Financial Trading Framework
#40I spent about an hour looking at the code and found some glaring issues that should be fixed before trusting it with real money. - Yahoo News is introduced twice (sentiment and news analysis) which double weights it - Sentiment analysis prompt primes the model to be bullish on Nvidia. - In the self learning loop there is a complex parsing bug that results in hallucinated memories when agents return truncated response…
There is also problem with underlying models. There was a study, where they always repest the same investing/management strategy: trust strangers, be open minded/adopt to new unproven ideas, prefer cooperation... Basically they were trained on disney-boomer bull(shit) market of last 15 years.
They have zero guards against market manioulations, and will get wiped without bull market!