Using AI to make trading decisions

If anyone has ever used AI ( artificial intelligence) based advisors to trade in stocks, please let me know.

1 Like

Interesting topic.

I’ve read those stories about teenagers making millions from their bedrooms doing just this.

But I guess if everyone did it, it wouldn’t work.

Not something I know anything about, but I’ll keep an eye on the responses :+1:

It doesn’t work, otherwise everyone would do it.
If someone sells to you a way to outperform the market, it’s always bogus. If they knew how to do that, they’d just trade themselves instead of making less money selling it to you.

There are big players that use ML in trading, but that is not accessible to small time investors and not a cash cow (yet) either afaik. The stock market is a complex system and it’s incredibly hard to model complex systems.
I’ve done some Qlearning trading (RL) on historical data for fun and it’s very hard and doesn’t perform that well on subsequent test transactions.

Trading generally does not work for retail investors. Barely anyone consistently and long-term outperforms buy&holding the market.

10 Likes

Perhaps the only plus side of allowing AI to make trading decisions is that it removes emotion from investing, ie you’re not trading because of FOMO, panicking etc.

2 Likes

Arguably AI is worse for stuff like this that a traditional algorithm. with machine learning you don’t know the reason it’s made the decision it’s made. With something like a traditional Expert Advisor algo at least you know the parameters and you can set a strategy

Not really, since the AI has an off button.

AI PhD here. AI is worse for another reason: amplification of bias in the data and / or bogus correlations. If anyone is considering using an AI assistant for that, firstly ask yourselves: What data has it been trained on? Is this data that can give me what I want? E.g. if it’s trained to predict performance based on historical performance, then this is just a case of “past performance is not predictive of future performance” but on steroids. If it’s trained additionally on things such as news articles, etc, then this is just hype/emotional trading on steroids. The only case in which I would consider using an algorithm to do my investments for me is if the algorithm is based on analysis of the fundamentals of the company. But you don’t really need AI for that.

4 Likes

Will the AI turn itself off?

Hey @mig

Interesting. My initial thought was that AI and ML may be capable of spotting correlations between general industry sectors that are too complex for the average fund to spot. What are your thoughts on this?

Was it not true that the institutions who first leveraged computers in finance and trading had a competitive advantage in the market and so their gains were increased, and if so, would leveraging AI in the right way offer a similar advantage?

Do you think that the market is too complex for AI to generate market signals to help steer trades? If not, what data do you think may help achieve success?

Great question @Indy123 and welcome.

No, humans will when they panic lol

… no one turned off Warren Buffet :wink:

1 Like

Good questions. It really comes down to two things: what do we mean by AI/ML and what data are we using to answer what question.

If we mean deep learning and very large models, those make sense only when we have stupid amounts of very high dimensional data. Think all of the internet (what ChatGPT is trained on). Models like that are crazy expensive both to train and to run. But let’s say we make them more economical. Does it make sense to use such a model for trading? I don’t know, I suppose the latest news articles will affect the market short term, but does it make sense to exaggerate daily bull and bear responses with ML? All of this leads to a feedback loop where the responses of the AI make it even more likely to respond like that in the future (buying something makes the price go up, etc). Somehow it doesn’t make sense to me, especially long term. But I suppose very short term, you are right: if only one player has a good large model for predicting daily trends from news articles, they may be able to do more successful day trading. Until others catch up. If this happens, I predict the market as a whole will become super crazy with lots of volatility for some time.

If you mean more established statistical and ML methods, those are already used all across finance. I’m not super familiar with the nuances here, but all sorts of regressions, time series modelling, probabilistic programming is definitely used, both for trading and for fundamentals / longer term strategy analysis.

I definitely don’t think the market is too complex for an AI to understand. But how we use data to answer questions is complex. It’s so easy to build a nonsense model that predicts nonsense but looks like it’s working. It reminds me of an example with the police in some city in the states (don’t quite remember which one). They build a model to “predict areas in the city where crime is likely to occur”. They trained it on data about where historically crime has been observed to occur: in other words, areas where the police was patrolling and observed crime. Of course, the police is more likely to patrol certain areas (e.g. black neighbourhoods) than others, so of course it will observe more crime there than in other areas. Having been trained and retrained on this data, the AI simply predicted more and more and more patrolling in those areas, which led to more and more policing of these areas. I hope it’s clear that what we wanted to predict is crime, but what we actually ended up predicting is where the police is patrolling. And on top of that we created a positive feedback loop. Something similar will happen if we implement trading AIs.

3 Likes

the man who solved the market

Thanks @mig. I agree on all points.

Thinking of feedback loops, they are apparent in AI art in a big way. It’s like the networks are on acid! Nb there is a theory that babies are in a similar state and my be why they are able to learn so rapidly; something to do with cranial pressure before the skull fuses (also see trepanning (not advised)).

I guess that the magic of the market is in the risk taking and trusting that in general the market will grow. It allows growth but there is some waste but the waste is needed just as much as the rest. If Ai were able to solve the market, I imagine that AI would be employed to re-garble it somewhat - just to maintain the system.

Fascinating area of enquiry. I will be watching this thread.

Link not working for me. “Something went wrong”.

this is the link “The BEST Trader In History - Legend Jim Simons - Strategy/Style/Story - YouTube

Ok. seems link is getting auto converted to associated text. Search Jim simons on youtube if the link doesn’t work for you