Sunday, April 27, 2008

DOE short-term forecasts analysis

In an article in May’s Energy Economics, a team of two U.S. professors and one energy-trading V.P. publish their findings on the accuracy of the DOE’s short-term supply forecasts. They looked at “one-step-ahead” forecasts of quarterly crude, natural gas, electricity, and coal supplies. Dr. Webber touched on forecasting in some earlier lectures, in a most negative manner. Although these authors spin the tale a bit more positively, you should hear them dance around any negative connotations. Some findings are posted below in “quotes”, with my translation in italics:

“Results suggest the U.S. Department of Energy (DOE) supply forecasts for U.S. domestic energy products are generally more accurate than a naïve alternative”

Our numbers show the DOE forecasts are usually better than a 5-year-old’s.

“Across the four markets examined—crude oil, natural gas, coal and electricity—there was not a particular sector where the EIA is clearly excelling or lacking.”

The EIA forecasts may, perhaps, be a waste of time.

“So, it would appear that there is a room for improvement in the crude oil supply forecasts.”

Please learn to do a better job in crude forecasting.

“Overall, the results shed a positive light on the forecasting ability of the EIA.”

The EIA forecasts are not completely useless like we had imagined.

“Indeed, the evidence presented here implies that the EIA generally does an admirable job of forecasting energy supplies at a one-quarter-horizon.”

We can reasonably say with some bit of certainty that the EIA forecasts are somewhat decent enough to rely on for budgeting for your summer vacation.

Source: Sanders, Manfredo and Boris, 2008. Accuracy and efficiency in the U.S. Department of Energy's short-term supply forecasts, Energy Economics 30 (2008), pp. 1192-1207.