Wine authenticity assessed via trimming

An authentic food is one that is what it claims to be. Consumers and\r\nfood processors need to be assured they receive exactly the specific product they pay\r\nfor. To ascertain varietal genuinity and distinguish doctored food, in this paper we\r\npropose to employ a robust mixture estimation method. It has been shown to be a\r\nvalid tool for food authenticity studies, when applied to food data with unobserved\r\nheterogeneity, to classify genuine wines and identify low proportions of observations\r\nwith different origins. Our methodology models the data as arising from a mixture of\r\nGaussian factors and employ a threshold on the multivariate density to bring apart the\r\nless plausible data under the fitted model. Simulation results assess the effectiveness\r\nof the proposed approach and yield very good misclassification rates when compared\r\nto analogous methods.