One simple way of predicating short-term stock price is to simply take the average price of the previous few days.

I wonder if there is some AI model/methods which can do better than this using only previous prices as input?

  • 3
    $\begingroup$ If it existed, it would probably be so widely used that any prediction, say a price rise in 5 minutes, would push up the price so that price increase disappears. See the efficient markets hypothesis. So any such AI model will be subjective and carry risk. $\endgroup$ Mar 14 at 8:13
  • $\begingroup$ Check the follow up! ai.stackexchange.com/questions/39594/… $\endgroup$
    – Rexcirus
    Mar 15 at 10:32

1 Answer 1


Attempt-wise, yes. All big players (e.g. J.P. Morgan) in the stock market who do automatic trading utilize reinforcement learning systems, which may contain components for short-term prediction.

But as @miguelmorin commented, if a simple algorithm ever exists, it would be so widely adopted until any gain from it disappear - that is, totally wrong predictions.

AFAIK, in long run, the only winners in short-term trading are those who can hire dozens of experts to build reinforcement systems, which are hosted on multi-million hardware, listening to thousands of market signals coming in from dark fibers 24/7, plus algorithms which can digest and place orders in less than 1 second (perhaps faster than our eyes seeing a price and our brain perceiving it). So, good luck!


You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .