Google stock markets are getting better and better at forecasting future events, according to research.
Researchers at the University of Illinois and the Massachusetts Institute of Technology used an algorithm that uses machine learning to predict the market’s next moves based on data from its trading algorithms.
They found the algorithm correctly predicted when a company’s shares were trading at an artificially high price and when it was trading at a lower price.
In addition, the researchers found that stock market predictions could be made on average two to three years ahead of actual events.
The team says it is the first time that artificial intelligence has been used to predict stock prices.
“The ability to accurately predict future events is important for financial markets,” said the paper’s lead author, Professor Daniel M. Katz.
“While the market has always had the ability to predict events, this work is the most sophisticated example yet of this capability in a human-controlled market.”
The team has used artificial intelligence to predict when stock prices will go up or down in the past.
In the past, the algorithm has been able to predict that the price of a stock would rise when it had a high correlation with a certain event, such as when an event such as the September 11 attacks would occur.
But in the latest research, the team was able to identify the best times to buy and sell stocks based on how their prices fluctuated during an event.
They then used that information to create an algorithm to predict how the market will react to the next stock event.
“It is the best method for predicting future stock market movements to date,” Katz said.
“What makes this method unique is that it is based on the knowledge of historical stock market fluctuations, and therefore allows for a predictive ability for future events.”
The researchers say their algorithm is capable of predicting how the price will change in different markets based on historical events.
“We are now able to use this method to predict a large fraction of the stock market’s future price movements,” said senior author Professor Michael J. Johnson, who is also a post-doctoral fellow in Katz’s lab.
“This is important as stock market analysis is highly dependent on stock market fundamentals and events in the world, such that it takes a long time to achieve the same results.”
The research, published in the journal Proceedings of the National Academy of Sciences, builds on work published last year that demonstrated the ability of artificial intelligence systems to predict future stock movements in a way that is nearly identical to the human-made stock market.
The researchers are now developing their own algorithm to accurately forecast future stock price movements, which they hope to be ready for use by the end of the year.