Wednesday, December 31, 2014

Forecasting

Two things usually occur in late December. Economists make economic forecasts for the coming year -- and others remind them how terrible their forecasts of the previous December have turned out. How about those oil prices?  Here are more from a year ago.

The economics we know helps us to grasp (perhaps forecast) broad patterns but it is not a good guide to the sorts of point forecasts that so many love to make. We are comfortable saying that if a Cuban entrepreneurial class is allowed to emerge and be free, the island will flourish.  Here are others of that nature.

But most of us have no clue what the price of gold or real estate or whatever, even the SP 500, will be at the end of next December. In their private lives, even economists buy index funds.

The WSJ's Jason Zweig discusses all this in his column today. He describes Ken Arrow's stint as weather forecaster during WWII.  Arrow saw that the models used for one-month weather forecasts were awful but was told by superiors, "The Commanding General is well aware that the forecasts are no good. However, he needs them for planning purposes." This all happened before Catch-22 was written.

Zweig also points us to a study that demonstrates the superiority of simple over fancy models. He also cites an on-going evaluation of forecasts, maintained by Phillip Tetlock and his associates.


In a WSJ op-ed of May 4, 1998 (I held on to it for great classroom discussions), Edward Yardeni wrote about Y2K, "We must prepare for the possible collapse of essential government services, including tax collection, welfare payments, national defense and air traffic control."  One can say that the alarmists perhaps helped us prepare -- and we did avoid the calamity.

Happy New Year.

ADDED

If you want to play, this is via Bill McBride.