Posted by: Tobias | August 26, 2008

Martingal, ick hör dir trapsen

Whenever the oil price moves by a few dollars in one direction or another or whenever the Dollar strengthens or weakens considerably against the Euro, Der Spiegel or another paper of comparable quality (ahem) will run an article on it and try to answer the pertinent W questions: What happened? Why did it happen? And where do we go from here?

I’m often fairly skeptical of the explanations given for why the price moved, because most of them just seem like ad-hoc rationalizations to me. But what drives me positively mad is that they never seem to have any trouble finding people who will claim to know authoritatively which way the price will move in the future.

Here’s a heuristic I’ve adopted for such people: If they claim to know which way exchange rates or oil prices will move they’re full of shit and anything else they say should be discounted accordingly

One of the most well-known pieces in the history of empirical finance is a 1983 paper by Richard Meese and Kenneth Rogoff which comes to the (admittedly somewhat startling) conclusion that none of the economic models of exchange rate determination (purchasing power parity, interest rate parity, etc.) can outperform a pure random walk model. A random walk model is one in which (roughly) the chance the exchange rate going up is exactly equal to the chance that it will go down at any point and over any period of time. And if this is the case, the best forecast of what the exchange rate will be tomorrow, a week, month or even a year from now is today’s exchange rate.

Of course since 1983 plenty of people have challenged the Meese-Rogoff conclusions and claimed to have found some predictability in exchange rate time series. But as far as I know, the basic result has largely stood the test of time.

And now comes the news that the same seems to hold for oil prices. Justin Wolfers reports:

I recently dug into the recent oil literature and discovered something amazing: It is easy to do better than the experts. At least this is what I learned from a recent working paper, “What Do We Learn From the Price of Crude Oil Futures?” by Michigan economists Ron Alquist and Lutz Kilian.

The authors compare a whole range of different ways of forecasting oil prices: they look up the Consensus Forecast (from a survey of expert economic forecasters), oil futures, the difference between the oil price in futures and spot markets, and also a range of more or less complicated econometric models that take account of recent trends, as well as variables like the interest rate.

And it turns out that they all do worse than one simple forecast: the current oil price. That’s right: the most accurate forecast of oil prices over the next month, year, or quarter is the current oil price. We call this the no-change forecast.

So, at least on these questions, just stop listening to the self-proclaimed experts.

(oh, and sorry about the bad pun in the title. Luckily only German-speaking statisticians will understand and bang their heads against the wall in response)



  1. The title is indeed fantastic! Also with the content of the post, I completely agree.

  2. Can’t comment on the title, but on the post:

    It is said after some many years of evolution and research we still have to accept that somethings can’t be modeled or predicted –> the pure random walk model or current price still outperforms. Maybe it is for the better anyway, after all what would happen if we could predict it?!

  3. Well, these things have some sort of feedback built into them. After all, if you could predict the exchange rate with some degree of accuracy, you could trade on the basis of that information thereby eliminating any further predictability. Lack of predictability therefore means (at least to a first approximation) that the markets are reasonably efficient, i.e. are incorporating all relevant information at all times.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s


%d bloggers like this: