AI That Picks Stocks Better Than the Pros "The ability to predict the stock market is, as any Wall Street quantitative trader (or quant) will tell you, a license to print money. So it should be of no small interest to anyone who likes money that a new system that works in a radically different way than previous automated trading schemes appears to be able to beat Wall Street's best quantitative mutual funds at their own game. It's called the Arizona Financial Text system, or AZFinText, and it works by ingesting large quantities of financial news stories (in initial tests, from Yahoo Finance) along with minute-by-minute stock price data, and then using the former to figure out how to predict the latter. Then it buys, or shorts, every stock it believes will move more than 1% of its current price in the next 20 minutes â and it never holds a stock for longer."
Really? Since when AI can really understand the information that is feeding in. Let's get serious There is NO such thing as real Artificial Intelligence, at least not now. Creative? Zero. What defines a real AI? > The ability to learn new things by itself and make proper connection. Reality for now?.. No! Just driven by some limited algorithm. Until then, let this "super" AI's come into the market to provide me better fills and provide liquidity>money to me. Regards, A humble manual trader.
It can only understand words, not sentences, but apparently that's enough. "The five verbs with the highest positive impact on stock prices are planted, announcing, front, smaller and crude." Of course once you know what the AI is looking for you can game the system to give your stock a boost. Did you hear? BP is announcing a smaller crude oil output. P.S. Did an AI write this article? I don't think those are all verbs.
The video mentions this towards the end: http://technology.timesonline.co.uk/tol/news/tech_and_web/article4742147.ece Reports of United Airlines bankruptcy caused stock to drop from $12 to $3 even though a human reading the story could have told from the content that the story was not current news. The "news" was from 2002 and was mislabeled as current (2008) when Google's news crawler couldn't figure out the date of the story.
In the '90's I did the same thing. I had software that popped up news stories according to key words. I had a database that would give me the annual sales of a company so I could guess the impact of a new contract for example.. the headline would pop up sorted according by company annual sales and other attributes. I can't recall all the details now but it was a lot fun but hectic work. Things like FDA Approval on a really small company would skyrocket an issue's price.. and you had about 30 seconds to press the buy button too... I have a library of all the stories that made money, it's a couple of feet of shelf space... there were bad trades of course but I concluded at that time that AI was not going to cut it until it could really understand a story somewhat the way a human would..