Here's how to predict future oil prices
Experts use complicated formulas to predict the future price of oil. But commentator Justin Wolfers says he's got a simple do-it-yourself method that works even better.
Justin Wolfers (Wharton School, Univ. of Pennsylvania)
More on Commentaries, Oil
TEXT OF COMMENTARY
Kai Ryssdal: When we talk about oil prices, it's usually a futures contract that we're talking about. An agreement to pay a pre-arranged price for a barrel of oil at some point in the future, hence future contract. The art of the deal, though -- and the way to make money -- is to guess that future price of the world's most actively-traded commodity. Commentator and economist Justin Wolfers thinks he's got it all figured out.
JUSTIN WOLFERS: Important decisions about our family finances -- things like which car to purchase or where to choose to live -- all of it hinges on whether today's high oil prices are here to stay, or whether this is just a temporary blip.
And there are dozens of talking heads on TV, pontificating about the latest oil industry developments.
But in fact, you're more of an expert than any of these talking heads. Or you will be, when I give you my secret forecasting formula.
Here it is: The single best forecast of oil prices in one month, three months, or a year is -- [sound of drumroll] -- today's oil price.
With oil prices at exorbitant levels, I'm forecasting that next year's price will also be at . . . exorbitant levels. I'm not saying that prices won't change, but I am saying they're about as likely to go up as they are to go down. Let's call this the no-change forecasting rule. It won't work for everything, but it does pretty well for oil prices.
In fact, Ron Alquist and Lutz Killian, two University of Michigan economists, recently assessed the forecasting performance of the no-change rule. Amazingly, this simple rule did better than the average of dozens of professional forecasters! In fact, the no-change forecast was 34 percent more accurate at predicting oil prices in three months' time, and 18 percent more accurate at predicting prices in a year's time. While professional prognosticators might argue that this difference isn't statistically significant, it sure is embarrassing.
Others ignore the professional forecasters and focus instead on what futures markets are saying. But it turns out that even futures prices are not as accurate as our simple formula. Even sophisticated econometric models don't yield better forecasts than our simple no-change rule.
The truth is that forecasting oil prices is so darn hard that complicated formulae add nothing but complexity. And so the simplest forecasting rule also turns out to be the best. Don't you wish that all of economics was this easy?
Ryssdal: Yes, we do. Justin Wolfers teaches at the University of Pennsylvania's Wharton School of Business.
Comments
Comment | Refresh
02/09/2009
The forecasting performance of the no-change rule works for just about everything. Including what time you'll get up tomorrow.
From Framingham, MA, 10/25/2008
It's ironic that this was concocted within two days of the peak price of oil on July 15, 2008.
Apparently, common sense and a sense of history were not a factor. It was a bubble, with self-fulfilling predictions about ever higher future oil prices. Like all bubbles, it had to burst. The author may have also been whistling in the dark.
With the coming demise of hedge funds and a new sense of reality (that may last until the next naive generation comes along), maybe we'll see no more bubbles for the next 30 to 40 years. And if bubbles are reborn after that interim, at least most of us won't have to be around to see them.
From Silver Spring, MD, 07/18/2008
Nathaniel,
a quick Google search provided this link to a copy of the paper from the researchers mentioned in the report. It does actually have some math showing their work :).
http://www.crei.cat/activities/sc_conferences/36/papers/killian.pdf
07/18/2008
It sounds like your commentator read an article or multiple articles. I'd like to see a reference to where the statistics in this commentary are coming from. I have no way to verify that what he has said is correct. Isn't it standard practice to include such references?
Post a Comment: Please be civil, brief and relevant.
Email addresses are never displayed, but they are required to confirm your comments. All comments are moderated. Marketplace reserves the right to edit any comments on this site and to read them on the air if they are extra-interesting. Please read the Comment Guidelines before posting.
You must be 13 or over to submit information to American Public Media. The information entered into this form will not be used to send unsolicited email and will not be sold to a third party. For more information see Terms and Conditions and Privacy Policy.