The researchers say their work can be used to create predictive models, which would help identify individuals at risk of panic selling.
“Financial advisors have long advised their clients to stay calm and weather any passing financial storm in their portfolios,” wrote Daniel Elkind, Kathryn Kaminski, Andrew Lo and colleagues. “Despite this, a percentage of investors tend to freak out and sell off a large portion of their risky assets.”
The extreme emotional swings of being a stock-market investor has long fascinated behavioral scientists. While the MIT study didn’t explore why exactly investors panic sell, the intense fear and desire to give up and get out of the market is well-known to any trader.
Countless studies have shown that people are better off staying put in a broad diversified portfolio, yet the promise of big riches and the terror of losing it all continues to drive frenetic trading patterns.
In the study, the researchers defined a panic sale as a plunge of 90% of a household account’s equity assets over the course of one month, of which 50% or more is due to trades.
“Panic sales are not random events,” the researchers wrote, saying its possible to identify clear trends in the data. They found that specific types of investors, such as those with less than $20,000 in their portfolio, also tend to liquidate more often.
“Subtle patterns in portfolio history, past market movements, and demographic profile can be exploited by deep neural networks to accurately predict if an investor will panic sell in the near future,” they added.
Photo by Maxim Hopman on Unsplash
