On Groundhog Day, Honoring A Forecasting Giant

This Groundhog Day I want to honor economic forecasters—and one in particular, Herman Stekler—rather than make fun of them, which is what I’ve tended to do on past Groundhog Days. Herman has had a 60-year career in forecasting and is still making predictions on everything that moves, including Super Bowl games. He recalls that the interview for his first job at Berkeley “occurred during the famous NY Giants–Baltimore Colts championship football game of 1959. I was a Giants fan, and when I left my hotel room they were ahead; I forecasted the final outcome incorrectly.”

Herman believes that forecasters should predict recessions early and often: “… the cost of a recession is so great that a forecaster should never miss one … Some people argue that turning points are unpredictable. I disagree. I have never had trouble predicting recessions. In fact, I have predicted n+x of the last n recessions.”

Herman’s colleagues and friends organized a conference on his 80th birthday and the proceedings have just been published in a special issue of the International Journal of Forecasting. The conference versions of the papers are available here.

The issue has an article by Fred Joutz, Tara Sinclair and me, which summarizes Herman’s extraordinary career—the early work on forecasting turning points and why forecasters seem to miss nearly every one of them; the first forecasting assessments of the Fed’s Greenbook forecasts; and much more.

Forecasting is difficult and I honor the people who have to do it. My own interest in the topic was triggered by my awful forecasts for growth in the Asian crisis economies in 1997-98. I have continued my own “astonishing record of complete failure” by completely failing to forecast the recent sharp decline in oil prices. I did call the Patriots-Seahawks outcome correctly.

Posted by at 2:21 PM

Labels: Forecasting Forum


Subscribe to: Posts