Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes

Experts give scores to wines, which are quality proxies for marketers and buyers. The production of scores proxying wine quality is explained by a set of observable objective attributes, plus another set of unobservable and subjective (sensory) features, and randomness. Is it possible to use a Stoch...

Full description

Bibliographic Details
Main Authors: Ferro, Gustavo Adolfo, Gatti, Nicolas
Format: info:ar-repo/semantics/artículo
Language:Inglés
Published: Taylor and Francis 2024
Subjects:
Online Access:http://hdl.handle.net/20.500.12123/20032
https://www.tandfonline.com/doi/10.1080/09571264.2024.2421548
https://doi.org/10.1080/09571264.2024.2421548
_version_ 1855038160691527680
author Ferro, Gustavo Adolfo
Gatti, Nicolas
author_browse Ferro, Gustavo Adolfo
Gatti, Nicolas
author_facet Ferro, Gustavo Adolfo
Gatti, Nicolas
author_sort Ferro, Gustavo Adolfo
collection INTA Digital
description Experts give scores to wines, which are quality proxies for marketers and buyers. The production of scores proxying wine quality is explained by a set of observable objective attributes, plus another set of unobservable and subjective (sensory) features, and randomness. Is it possible to use a Stochastic Frontier Approach (SFA) to understand whether objective and subjective (sensory) characteristics of wines explain the differences in ‘producing’ wine scores? We estimate a wine quality stochastic frontier production function, using a database of 1800 top-scored wines, in 18 years-window encompassing objective determinants that largely explain the scores (price, production, year, grape, country, etcetera), being aware that sensory aspects related to wine grading are unobservable. We find that the observable variables explain part of the ‘efficiency’ in attaining scores and suggest that ‘inefficiency’ in score generation can be attributed to unobservable (sensory) variables.
format info:ar-repo/semantics/artículo
id INTA20032
institution Instituto Nacional de Tecnología Agropecuaria (INTA -Argentina)
language Inglés
publishDate 2024
publishDateRange 2024
publishDateSort 2024
publisher Taylor and Francis
publisherStr Taylor and Francis
record_format dspace
spelling INTA200322024-11-01T10:20:26Z Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes Ferro, Gustavo Adolfo Gatti, Nicolas Vinos Calidad Modelos Estocásticos Wines Quality Stochastic Models Stochastic Frontier Approach Stochastic Frontier Analysis Fronteras Estocásticas Experts give scores to wines, which are quality proxies for marketers and buyers. The production of scores proxying wine quality is explained by a set of observable objective attributes, plus another set of unobservable and subjective (sensory) features, and randomness. Is it possible to use a Stochastic Frontier Approach (SFA) to understand whether objective and subjective (sensory) characteristics of wines explain the differences in ‘producing’ wine scores? We estimate a wine quality stochastic frontier production function, using a database of 1800 top-scored wines, in 18 years-window encompassing objective determinants that largely explain the scores (price, production, year, grape, country, etcetera), being aware that sensory aspects related to wine grading are unobservable. We find that the observable variables explain part of the ‘efficiency’ in attaining scores and suggest that ‘inefficiency’ in score generation can be attributed to unobservable (sensory) variables. Los expertos califican los vinos para ayudar a los vendedores y compradores a identificar su calidad. La calidad del vino se determina por atributos objetivos observables, como también por características no observables y subjetivas (sensoriales) y aleatorias. ¿Se puede usar la técnica de frontera estocástica para analizar si las características objetivas y subjetivas (sensoriales) de los vinos explican las diferencias en las puntuaciones que reciben? En este estudio, se calculó una función de frontera estocástica de producción de la calidad del vino utilizando información sobre 1800 vinos de alta puntuación que abarca una ventana de 18 años. Se consideraron factores como el precio, la producción, el año, la uva, el país, etc., aunque se reconoce que los aspectos sensoriales no se pueden medir directamente. Encontramos que las variables observables contribuyen a la “eficiencia” en la obtención de puntajes, mientras que la ”ineficiencia” en la generación de puntajes puede deberse a variables no observables (sensoriales). Centro de Investigación en Economía y Prospectiva Fil: Ferro, Gustavo Adolfo. Universidad del Centro de Estudios Macroeconómicos de Argentina (UCEMA); Argentina Fil: Ferro, Gustavo Adolfo. Consejo Nacional de Investigaciones Científicas y Técnicas; Argentina Fil: Gatti, Nicolás. Instituto Nacional de Tecnología Agropecuaria (INTA). Centro de Investigación en Economía y Prospectiva (CIEP); Argentina Fil: Gatti, Nicolás. Universidad del Centro de Estudios Macroeconómicos de Argentina (UCEMA); Argentina 2024-11-01T10:11:57Z 2024-11-01T10:11:57Z 2024-10 info:ar-repo/semantics/artículo info:eu-repo/semantics/article info:eu-repo/semantics/publishedVersion http://hdl.handle.net/20.500.12123/20032 https://www.tandfonline.com/doi/10.1080/09571264.2024.2421548 Ferro, G., & Gatti, N. (2024). Explaining wine scores through stochastic frontier analysis. An intent to measure sensory non-observable attributes. Journal of Wine Research, 1–20. https://doi.org/10.1080/09571264.2024.2421548 1469-9672 0957-1264 https://doi.org/10.1080/09571264.2024.2421548 eng info:eu-repo/semantics/restrictedAccess http://creativecommons.org/licenses/by-nc-sa/4.0/ Creative Commons Attribution-NonCommercial-ShareAlike 4.0 International (CC BY-NC-SA 4.0) application/pdf Taylor and Francis Journal of Wine Research : 1-20. (Published online: 28 Oct 2024)
spellingShingle Vinos
Calidad
Modelos Estocásticos
Wines
Quality
Stochastic Models
Stochastic Frontier Approach
Stochastic Frontier Analysis
Fronteras Estocásticas
Ferro, Gustavo Adolfo
Gatti, Nicolas
Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes
title Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes
title_full Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes
title_fullStr Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes
title_full_unstemmed Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes
title_short Explaining wine scores through stochastic frontier analysis: an intent to measure sensory non-observable attributes
title_sort explaining wine scores through stochastic frontier analysis an intent to measure sensory non observable attributes
topic Vinos
Calidad
Modelos Estocásticos
Wines
Quality
Stochastic Models
Stochastic Frontier Approach
Stochastic Frontier Analysis
Fronteras Estocásticas
url http://hdl.handle.net/20.500.12123/20032
https://www.tandfonline.com/doi/10.1080/09571264.2024.2421548
https://doi.org/10.1080/09571264.2024.2421548
work_keys_str_mv AT ferrogustavoadolfo explainingwinescoresthroughstochasticfrontieranalysisanintenttomeasuresensorynonobservableattributes
AT gattinicolas explainingwinescoresthroughstochasticfrontieranalysisanintenttomeasuresensorynonobservableattributes