Quants and quaffs
“ALTERNATIVE assets” suggests a class that offers something to enjoy. Philatelists delight in rare stamps; petrol-heads in classic cars; and oenophiles in that most liquid of assets, fine wine. The wine futures market, though, is pretty inefficient. Prices hinge on tastings of stuff that is still in the barrel, long before its reaches its fullest bloom. This en primeur pricing happens on experts’ palates, not in the equations of quants. Tristan Fletcher, of University College London, is among those who would like to change that, using some of the most probing equations computer science has yet devised.
There have been a few attempts over the years to tame the fickle wine market into an equation. These have relied on using what are known as linear regression models to make a palatable blend of facts about a given vintage out of particulars of the weather that year, the vineyard’s history of medallion-winning and so on. Linear regression takes the unseemly spray of these data points and draws through them the straight line that, over the course of time, has most closely approximated the price. Pick the point on this line where a particular vintage lies, and out comes a price prediction.
Such efforts have had mixed results, and Dr Fletcher thought he could be better. Instead of regression, he applied a form of artificial intelligence called machine learning. This is able to ferret out correlations (perhaps a great many of them, some weak or transient) that standard regression models gloss over. Rather than a simple straight line, the result is a price curve that snakes through the data, thus yielding, if the particulars of the calculation have been set up properly, stronger predictions than regression can manage.
Although it is known among investors in more standard asset classes, Dr Fletcher and his colleagues wanted to bring machine learning to bear on the wine market. They started with wines in the Liv-ex 100 (a kind of fine-wine FTSE) and looked only at the historical data on prices. They first ran an “autocorrelation test”—a way to quantify how calm or unruly prices had been in the past. As they report this week in the Journal of Wine Economics, they found two distinct groups. Half seemed to fluctuate over short periods, returning toward the mean price quickly. The other half seemed to trend up and down more wildly. The team then ran two types of machine-learning algorithms on these groups separately. For the wines of the calmer group (in which the market was, presumably, behaving efficiently), these algorithms outperformed the regression method by only a little. For those with more untamed price histories, though—those on which more money could, in principle, be made—machine learning roundly won.
Source: Quants and quaffs
Via: Google Alerts for AI