Predicting the stock market from photographs in the financial-press (new study)

If you’re inclined to make bets on the stock market, you’ll be at a considerable disadvantage if you don’t have a tight grip on so called ‘investor sentiment’ – in other words a reliable way of being able to judge the current ‘mood’ of the market. But, in practice, it’s notoriously difficult to gauge accurately. A team from Missouri State University, US, have come up with a novel method. Why not just analyse photos in the financial press to see if people look happy (or not) ? They’re calling it ‘Photo Sentiment’.

The researchers say that their machine-learning algorithms, which analysed photos in The Economist (that had been published over a period of 10 years) out-performed other more traditional methods by 2.6%.

“Photo Sentiment is found to predict short-term return reversal, trading volume and market volatility, and explain flows between equity and money market funds. A trading strategy based on Photo Sentiment outperforms the market index by 2.6% on an annual basis while assuming less risk.”

See: Obaid, Khaled and Pukthuanthong, Kuntara, A Picture Is Worth a Thousand Words: Market Sentiment From Photos (December, 2018). Available at SSRN

_{The Economist cover above is from July 1998}