Airfare forecasting has never been an exact science, but at least economists are trying to make it more precise. From the Observer:

An economist, Makoto Watanabe, has calculated that the optimum time to buy an airline ticket is eight weeks in advance of flying.

His yet-to-be-published findings also suggests that airline tickets are cheaper when purchased in the afternoons, rather than the mornings, prompting him to speculate that airlines are assuming business travellers will book their tickets at work in the morning on the company account, whereas leisure travellers are more likely to book from home in the afternoon.

The eight-week result stems from work published in the latest edition of the Economic Journal in which Watanabe and his colleague, Marc Möller, offer intimidating equations such as ?A = gUG + min(k – g, (1 – g)(1 – r)) as part of the complex formula, where ? equals profit, that determines advance ticket purchases.

The eight-weeks-in-advance rule is plausible, though even then it’s more likely a rule of thumb than a rule as such.

It’s also potentially geographically limited in its validity: If the study was based on flights originating in London, then eight weeks may be the sweet spot for London travelers. But not necessarily for every market.

The buy-later-in-the-day guidance may be more problematic. Recall the myth that buying Wednesday at midnight is the best time to buy airline tickets? This is a variation on that theme.

Finding the cheapest fare is much like picking the precise bottom for a stock. You’re better off trying to buy cheap (I like the email alerts from FareCompare.com for finding fares that are below the norm), not trying to pick the bottom.

(hat tip to Budget Travel)