In the latest issue of Forbes magazine, Validea CEO John Reese examines why human beings tend to exercise bad judgment when making investment decisions, and how using quantitative investment strategies can help them overcome their poor predictive powers.
Reese points to a study that Philip Tetlock detailed in his 2005 book, Expert Political Judgment: How Good Is It? How Can We Know? Tetlock’s seven-year study found that even the best human forecasters were accurate only about 20% of the time when predicting political and economic events. Crude algorithms, however, were able to predict 25% to 30% of total variability in outcomes and more sophisticated models predicted 47% of variability — more than twice as accurate as human experts.
“Our own humanity makes us bad forecasters,” Reese writes, citing more data that shows investors tend to jump in and out of stocks at very inopportune times. “We are emotional, biased creatures. Models, on the other hand, stick only to the cold, hard facts. They aren’t perfect, of course, but in the stock market they don’t need to be. If you are right on a little over half of your picks, you will make nice money.”
By using quantitative strategies when investing, Reese says investors can help keep emotion at bay and focus on the facts. He discusses his own quantitative “Guru Strategies”, each of which is based on the approach of a different investing great, and offers three of their recent picks. Among them: Curtiss-Wright, which gets high marks from his Benjamin Graham-inspired strategy.