Forecasting oil futures by Justin Wolfers (Wharton School, Univ. of Pennsylvania) on Marketplace (a great show by the way)
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.
This is another example of how tricky it is to predict financial markets. I am a bit surprised for relatively longer periods (like a year) the professionals do so poorly. My father, a statistician (among other things), challenged me to predict the movement of stocks on a daily basis better than his prediction (which was no change). I can’t remember the result – which makes me think I failed. I think I would be more likely to remember if I succeeded.
Related: Prediction Markets at Google – Illusion of Explanatory Depth – 30 Year Fixed Mortgage Rates Graph – Randomization in Sports
Comments
1 Comment so far
Oil has fallen to $40 a barrel from nearly $140 less than a year ago. Now that $140 level was the result of a huge spike in the price. But if I owned a bunch of oil (as a country or a company) I sure wouldn’t want to sell it at $40…