Soybean production value in the Rio Grande do Sul under the GAMLSS framework
- 8 December 2020
- journal article
- research article
- Published by Taylor & Francis in Communications in Statistics: Case Studies, Data Analysis and Applications
- Vol. 7 (2) , 146-165
Abstract
In this article, we consider the more recent soybean production data in Rio Grande do Sul (years 2017 and 2018) and obtain regression models through the generalized additive models for location, scale, and shape (GAMLSS) approach, and provide a dashboard as a visualization tool of the considered variables. Two models are applied to explain and predict the soybean production value as a function of the covariates, such as produced quantity, number of establishments, and average yield in each city of RS. Validation and cross-validation methods are considered to assess whether the predictions provided by the fitted models are reliable. The fitted model with data of 2017 provides the best predictions. The GAMLSS framework may be more accurate than linear regression to model data related to soybean production, constituting them in a reliable and useful source to auxiliary the farmers and economic sector managers in making decisions.Keywords
This publication has 20 references indexed in Scilit:
- Indicadores Climáticos e a Produtividade de Soja no Rio Grande do SulRevista Brasileira de Meteorologia, 2019
- The role of soybean production as an underlying driver of deforestation in the South American ChacoGlobal Environmental Change, 2017
- Non-nested hypothesis testing inference for GAMLSS modelsJournal of Statistical Computation and Simulation, 2016
- An Introduction to Statistical Learning2013
- Soybean yield and quality a function oflime and gypsum applicationsScientia Agricola, 2006
- Soy Applications in Food2005
- Indicadores de produção de soja no Rio Grande do Sul comparados ao zoneamento agrícolaPesquisa Agropecuária Brasileira, 2004
- Soybean yield estimation by an agrometeorological model in a GISScientia Agricola, 2003
- Smoothing reference centile curves: The lms method and penalized likelihoodStatistics in Medicine, 1992
- Generalized Linear ModelsJournal of the Royal Statistical Society. Series A (General), 1972