Please use this identifier to cite or link to this item:
Title: Análise espacial da produção leiteira usando um modelo autoregressivo condicional
Other Titles: Spatial analysis of the dairy yield using a conditional autoregressive model
Keywords: Milk production
Linear regression
Spatial analysis
Conditional auto-regressive model
Produção de leite
Regressão linear
Análise espacial
Modelo auto-regressivo condicional
Issue Date: Apr-2010
Publisher: Universidade Estadual de Londrina
Citation: PONCIANO, P. F.; SCALON, J. D. Análise espacial da produção leiteira usando um modelo autoregressivo condicional. Semina: Ciências Agrárias, Londrina, v. 31, n. 2, p. 487-496, abr./jun. 2010.
Abstract: The dairy yield is one of the most important activities for the Brazilian economy and the use of statistical models may improve the decision making in this productive sector. The aim of this paper was to compare the performance of both the traditional linear regression model and the spatial regression model called conditional autoregressive (CAR) to explain how some covariates may contribute for the dairy yield. This work used a database on dairy yield supplied by the Brazilian Institute of Geography and Statistics (IBGE) and another database on geographical information of the state of Minas Gerais provided by the Integrated Program of Technological Use of Geographical Information (GEOMINAS). The results showed the superiority of the CAR model over the traditional linear regression model to explain the dairy yield. The CAR model allowed the identification of two different spatial clusters of counties related to the dairy yield in the state of Minas Gerais. The first cluster represents the region where one observes the biggest levels of dairy yield. It is formed by the counties of the Triângulo Mineiro. The second cluster is formed by the northern counties of the state that present the lesser levels of dairy yield.
Appears in Collections:DEX - Artigos publicados em periódicos

This item is licensed under a Creative Commons License Creative Commons