Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Financial market applications of LLMs

thegradient.pub

Post reply on HN
Print view View original

101–110 of 116 posts

Re: Financial market applications of LLMs

#101

Post by pid-1 »

richrichie wrote: post #72
conorh wrote: post #26

We are working on a project for a client which functions as an analysis tool for stocks using LLMs. Ingesting 10ks, presentations, news, etc. and doing comparative analysis and other reports. It works great, but one of the things we have learned (and it makes sense) is that traceability of the information for financial professionals is very important - where did the facts and information come from in what the AI is p…

I worked on a similar application and eventually we shelved it. We just could not be confident enough that the numbers in the report produced are correct. There were enough instances of inaccuracies to not use it for important decision making. Which actually meant a lot of double work.

Same experience here.

Re: Financial market applications of LLMs

#102

Post by ysofunny »

tiborsaas wrote: post #84
nexuist wrote: post #82

Earlier quoted context omitted.

I've sort of come around on this. Yes, everything you listed is valuable and good. But the reality is all of it was built with money that came from banks and investors. The only reason to do anything scientific is to get investors to give you money. If you do something scientific that does not make people want to give you money you will impact no lives. In this way gaming financial markets is indeed the only point to…

Take particle physics for example, the LHC was incredibly expensive and most people's life won't be better that that we've found the Higgs field. It was also not paid by investors but public money. There are quite a lot of science that's basic research and it's done for scientific curiosity only with no clear way of translating that to marketable applications.

> "most people's life won't be better that that we've found the Higgs field."

on one hand, I feel you should append "...yet" into that thought. just because it hasn't been useful technologically now doesn't mean it never will.

on the other hand, I've seen some physicists doubt that quarks even exist out of a controversy about Feynman's allegedly missing diagrams

Re: Financial market applications of LLMs

#103

Post by swexbe »

fragmede wrote: post #45
__MatrixMan__ wrote: post #14

Earlier quoted context omitted.

That should be the goal, right? Good ideas get the funding they need as if by magic, yet nobody is sitting on the sidelines collecting rent. The best thing that AI can do for finance is eliminate it.

that sounds really bad for everyone collecting rent

It would be great for people collecting rent. No fees, just profit.

Re: Financial market applications of LLMs

#104

Post by mattew »

jsemrau wrote: post #2

A lot of words for not bringing much new content to the discussion. I think the most interesting application of LLMs in Finance are (1) synthetic data models for data cleansing, (2) journal management, (3) anomaly tracking, (4) critiquing investments All of this should be done by professionals and nothing is "retail" ready.

This makes sense. Can you clarify what you mean by journal management in this context?

Re: Financial market applications of LLMs

#105

Post by btbuildem »

bobbruno wrote: post #90
blitzar wrote: post #86

Earlier quoted context omitted.

I have tested this with multiple veterans and none could tell them apart - but they had a high conviction on which random walks were a good buy and which were compelling shorts.

And there is your arbitrage opportunity. If you can model how analysts will react to a particular timeseries, even if it was random until that point, you have some information about the future. It'd be a good question to figure out if there is a consensus or majority about how to interpret patterns among the people making decisions or writing quant algos, that's something one could use.

That's an interesting take! You show them meaningless data, in order to extract their overall market sentiment, and use _that_ to inform your investing strategy?

Re: Financial market applications of LLMs

#106

Post by ahzhou »

nexuist wrote: post #82
jonahx wrote: post #20

Earlier quoted context omitted.

> is that gaming financial markets is the only real application of anything scientific medicine (living longer, curing disease, vaccines, etc), cheaper energy, cheaper transportation, cheaper construction, cheaper food, better communication, new forms of entertainment, just off the top of my head.

I've sort of come around on this. Yes, everything you listed is valuable and good. But the reality is all of it was built with money that came from banks and investors. The only reason to do anything scientific is to get investors to give you money. If you do something scientific that does not make people want to give you money you will impact no lives. In this way gaming financial markets is indeed the only point to…

If you’re saying that economics is a foundational driver of progress, then yes - almost by definition.

Banks and investors provide liquidity to the system, which is just one of many things the market demands.

Re: Financial market applications of LLMs

#107

Post by btbuildem »

nostrademons wrote: post #32
btbuildem wrote: post #30

There were some developments using LLMs in the timeseries domain which caught my attention. I toyed with the Chronos forecasting toolkit [1], and the results were predictably off by wild margins [2] What really caught my eye though was the "feel" of the predicted timeseries -- this is the first time I've seen synthetic timeseries that look like the real thing. Stock charts have a certain quality to them, once you've…

I used to work in financial software, and when writing the charting UIs, I'd wire them up to a randomwalk to generate fake time series data. It was a relatively common occurrence for a VP or the company CEO to walk by, look at my screen, and say "What stock is that? Looks interesting." Unpopular opinion backed up by experience: a randomwalk is the most effective model for generating timeseries that have the "feel" of…

I'd love to see some examples, if you have old screenshots laying around!

Your take conflicts with my toy hypothesis, and I wouldn't mind being proven wrong if it saves me time and effort.

I wonder if the folks who were fooled by your screens were fooled by the random data itself, or the fact that it was presented within all the familiar chrome and doodads that people associate with stock price visualization.

Re: Financial market applications of LLMs

#108

Post by __MatrixMan__ »

swexbe wrote: post #103
fragmede wrote: post #45

Earlier quoted context omitted.

that sounds really bad for everyone collecting rent

It would be great for people collecting rent. No fees, just profit.

Where would the profits come from? Any AI extracting them would be at a disadvantage compared to the equivalent AI that makes the same decisions otherwise but doesn't extract those profits.

Profits motivate the investor, but they impede the investment, and what we want are successful investments not happy investors.

I don't think human investors can manage to be less greedy than an AI designed to not be greedy. AI will get more efficient, while humans still have to eat. Also, the assumption we're working under is that humans also make worse investment decisions than the AI.

So if you're in need of abstractions to motivate others to help you with some venture that you can't do alone, why would you (or your employees) prefer the ones that have greedy third parties in the loop who are also misallocating resources?

There are domains where that human touch means something. We should not let AI run everything. But finance is not one of them, so why resist it becoming a solved problem?

The investors can compete on who can take smaller and smaller profits, aided by the AI, and once they've got their system nearly perfect, we copy it and have it take no profits at all. Thanks capitalism, you've done your job, now it's time to go get a different one.

If we start getting outcomes that we don't like, we can always just turn our backs on the AI-begotten abstractions and let the humans take another crack at it, but a system that runs itself without owners extracting profits should absolutely be the goal.

Re: Financial market applications of LLMs

#109

Post by scrollbar »

cpursley wrote: post #36
conorh wrote: post #26

We are working on a project for a client which functions as an analysis tool for stocks using LLMs. Ingesting 10ks, presentations, news, etc. and doing comparative analysis and other reports. It works great, but one of the things we have learned (and it makes sense) is that traceability of the information for financial professionals is very important - where did the facts and information come from in what the AI is p…

I assume you're ingesting PDFs. If so, how are you handling tables accurately?

A coworker presented a demo the other day of this - asking LLM (I think it was OpenAI) to extract the text from a PDF - each page of the PDF passed as an image. It was able to take a table and turn it into a hierarchical representation of the data (ie. Column with bullets under it for each row, then next column, etc.)

If you haven't tried maybe worth a shot

Re: Financial market applications of LLMs

#110

Post by coastermug »

cpursley wrote: post #36
conorh wrote: post #26

We are working on a project for a client which functions as an analysis tool for stocks using LLMs. Ingesting 10ks, presentations, news, etc. and doing comparative analysis and other reports. It works great, but one of the things we have learned (and it makes sense) is that traceability of the information for financial professionals is very important - where did the facts and information come from in what the AI is p…

I assume you're ingesting PDFs. If so, how are you handling tables accurately?

AWS textract now has the functionality to offer a table cell based on a query - if I’m not mistaken. I’ve seen nothing similar to this and would be very interested if there are other solutions.
Post reply on HN