FORECASTING AND PREDICTING

“It’s tough to make predictions, especially about the future.” Yogi Berra

____________________

The late Harvard economist John Kenneth Galbraith opined that “the only function of economic forecasting is to make astrology look respectable.” Yet equine businesses must forecast to develop operating plans.

The long-term accuracy of most economic forecasts is mediocre. For example, The Blue Chip Economic Indicators is an oft-quoted prediction of where the U. S. economy is headed, based on a survey of about 50 prominent economists. Barron’s characterizes the results: “Its track record is not encouraging.”

Similarly, hedge-fund managers charge very high fees to affluent clients, ostensibly to deliver alpha by predicting and selecting the best investments to make. But a popular measure of hedge-fund returns, the HFRX, reveals that the S& P 500 has topped average hedge-fund returns for nine of the past ten years. In 2012, nearly 9 of 10 hedge-fund managers lagged the S&P 500, which returned 16% versus 2.69% for hedge funds.

Agreement among forecasters is apt to reflect a herding bias. A case in point is pari-mutuel wagering on horse racing, where the favorite wins only about one-third of the time. Though consensus forecasts tend to be better than the vast majority of individual predictions, they are frequently not accurate enough to be helpful.

Forecasters are particularly susceptible to herding in situations where they have a common interest that may engender and reinforce a confirmation bias. Being in physical proximity—such as in Lexington or Ocala–can magnify the effect. Some investment companies purposely locate away from Wall Street for this reason.

Ray Dalio, the famed founder of the investment firm Bridgewater Associates, says: “The consensus is often wrong, so I have to be an independent thinker. To make any money, you have to be right when they’re wrong.”

Statistician Nate Silver, author of a 2012 book on forecasting advises: “…seek out forecasts that couch their predictions in percentage or probabilistic terms. They are a more honest representation of the limits of our predictive abilities.”

Initiate the forecasting task for a horse-racing business by representing several plausible scenarios (at least include best, worst, and most likely) and assigning probabilities to them. Then, as time goes by and further information becomes available about business conditions, revise both the probabilities and the appropriate tactical responses. This hedging of bets militates against a major cause of errant forecasting—overconfidence in the prospects for a single outcome.

Copyright © 2013 the Blood-Horse. Used with permission.