How 10 news photos can predict stock market movements, Australia study claims – 9News

This post was originally published on this site

Want the good news or bad news?

Turns out artificial intelligence (AI) can answer that very question if there’s a news photograph it can carefully scan and analyse.
And, according to a recently published RMIT University study, the AI’s answer to that question can help traders predict the mood of global stock markets and, in doing so, make better returns.
Using machine learning, the algorithm produces a daily score based on the types of photos used in global news reports. (Unsplash)

The system works, the study said, through the teaching of AI to analyse a list of the most popular news photographs on Getty, an image website used by media outlets around the world.

Getty’s top-10 list is made up of the most downloaded news photographs in the past 24 hours, which are submitted by photographers covering and capturing the world’s most important events of the day.

Using machine learning to interpret what is happening in each photograph, the AI will then calculate a “photo pessimism” score for that day, according to the study’s lead author Dr Angel Zhong.

Once the photo pessimism measure is delivered, Dr Zhong said the AI was then able to predict with “95 per cent accuracy” if a stock market will move up or down.

The top news photographs on April 16, 2003 – a day when ‘bad news’ prevailed. (Getty)

Dr Zhong said traders could use the pessimism score to invest in exchange-traded funds (ETF) which mirror the performance of stock market benchmarks such as Australia’s S&P/ASX 200 Index, or the S&P 500 and the Dow Jones Industrial Average in the US.

The study, Photo Sentiment and Stock Returns around the World, found that when people are in a bad mood and facing higher uncertainty, trends showed they tend to buy and sell more impulsively and intensively.

Dr Zhong said the pessimism score was “a good tool” which allows traders to get a snapshot of a market’s mood in a “very short time”.

The RMIT University study harnessed the powers of machine learning to scan an ocean of news photographs between 1995 and 2018, which were then applied to 37 stock markets around the world, including those located in developed and developing nations.

Machine learning and AI is already used by traders and companies to scan newspaper text but this method cannot easily overcome language barriers, meaning any analysis is often only a measure of the English-speaking developed world.

Dr Zhong said her study had developed a method which “transcended language barriers” so that the mood of global markets where all manner of foreign languages are spoken can be quickly gauged.

Research shows how investors can use news photos to better predict daily stock market returns. (Supplied)
The RMIT University study built further on work carried out by University of Missouri researchers who earlier this year devised a way to index daily investor sentiment using news photos in the US.

Dr Zhong said her study had trained the algorithm to recognise what constitutes good and bad news coverage in 37 markets, increasing the complexity and global accuracy of the University of Missouri research.

The pessimism score would often allow “predictability” of a market’s mood and movements for up to 72 hours, she said.

“If it is good today, what you can do is buy the ETF index funds that check the ASX and (there is a) 95 per cent chance that they would go up for the next few days.”

In trying to establish if a photograph was depicting good or bad news, the AI looked for clues such as body gestures, certain types of facial expressions, colours and subjects in the image.

“If you have guns and weapons, these tend to be related to war and this would indicate bad news,” Dr Zhong said, while “blue sky, children smiling and families together” were more likely good news.

No trading houses or investment funds have enquired about the AI technology yet, Dr Zhong said.

The information provided on this website is general in nature only and does not constitute personal financial advice. The information has been prepared without taking into account your personal objectives, financial situation or needs. Before acting on any information on this website you should consider the appropriateness of the information having regard to your objectives, financial situation and needs.

This post was originally published on *this site*