Component Analysis” (Barth et al., 2021). The authors want to produce an efficient classifier to shed light on the important features of wines in the classification. They draw on a dataset of 178 wines with various chemical measurements (alcohol, malic acid, ash levels, alkalinity of ash, magnesium levels, total phenols, flavonoids phenols, non-flavonoid phenols, proanthocyanins, color intensity, hue, OD280/ OD315, and proline) made from the varieties Nebbiolo, Grignolino, and Barbera. They employ principal component analysis (PCA) in the k-nearest neighbor (kNN) classification to deal with multicollinearity issues among the explanatory var- iables. PCA can identify the underlying dominant features and provide a more suc- cinct and straightforward summary over the correlated covariates. The study suggests that a combination of kNN and PCA yields a simpler and more interpret- able classifier than kNN based on all 13 variables. “Our final classifier is based on only two principal components, which can be interpreted as strength of taste and level of alcohol and fermentation in wines, respectively” (p. 56).
Ursula Landazuri-Tveteraas, Frank Asche, and Hans-Martin Straume analyze the “Dynamics of Buyer-Seller Relations in Norwegian Wine Imports” (Landazuri- Tveteraas, Asche, and Straume, 2021). Drawing on specific Norwegian wine import- ing companies and their foreign partners, the authors analyze how the duration of firm-to-firm trade relationships are affected by different factors in an 11-year period stretching from 2004 to 2014. The study finds that most trade relationships are short-lived, that is, more than 75% of trade relationships end after less than two years. In addition, the authors’ findings suggest a positive relationship between wine price and trade partnership durations. “Deeper firm-to-firm trade rela- tionships for more exclusive wines are likely due to higher search costs for high- quality products” (p. 83).
In the last paper of this issue, entitled “Do Whisky Investors Read the Bible? The Effect of Expert Ratings on the Vintage Single Malt Secondary Market,” David Moroz and Bruno Pecchioli find that quality scores of Jim Murray’s Whisky Bible are not powerful price predictors for whiskey ask prices (Moroz and Pecchioli, 2021). Their findings are similar to those in the wine literature (e.g., Ashenfelter and Jones, 2013) “in the sense that expert ratings are not efficient in explaining the price and can be improved. We show that distillery and bottler name, as a proxy for their respective reputations, are more accurate than ratings” (p. 99).
Karl Storchmann New York University
References
Alston, J., Arvik, T., Hart, J., and Lapsley, J. (2021). Brettanomics I: The cost of Brettanomyces in California wine production. Journal of Wine Economics, 16(1), 4–31.
Ashenfelter, O., and Jones, G. (2013). The demand for expert opinion: Bordeaux wine. Journal of Wine Economics, 8(3), 285–293.
Barth, J., Katumullage, D., Yang, C., and Cao, J. (2021). Classification of wines using princi- pal component analysis. Journal of Wine Economics, 16(1), 56–67.
Doan, M. P., and Sercu, P. (2021). Merging one’s way to the top: AB Inbev v. Heineken. Journal of Wine Economics, 16(1), 32–55.
Landazuri-Tveteraas, U., Asche, F., and Straume, H.-M. (2021). Dynamics of buyer-seller relations in Norwegian wine imports. Journal of Wine Economics, 16(1), 68–85.
Moroz, D., and Pecchioli, B. (2021). Do whisky investors read the Bible? The effect of expert ratings on the vintage single malt secondary market. Journal of Wine Economics, 16(1), 86–101.