Verossimilhança na seleção de modelos para predição espacial

Nesi, Cristiano Nunes; Ribeiro, Andressa; Bonat, Wagner Hugo; Ribeiro, Paulo Justiniano

doi:10.1590/s0100-06832013000200006

Cited by 5 publications

(3 citation statements)

References 20 publications

(21 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…One of the methods used in such characterization is geostatistics, which takes soil spatial variation patterns into account and provides techniques that enable construction of maps associated with one or more soil chemical properties (Cressie, 2015).One of the different methods used in geostatistical studies is spatial linear models, which have been widely evaluated by assuming a Gaussian stochastic process (Uribe-Opazo et al, 2012;Grzegozewski et al, 2013;Nesi et al, 2013;De Bastiani et al, 2015). This modeling enables estimation of spatial dependence parameters through the maximum likelihood method (ML), making inferential studies possible.…”

Section: Introductionmentioning

confidence: 99%

Slash Spatial Linear Modeling: Soybean Yield Variability as a Function of Soil Chemical Properties

Fagundes

Uribe-Opazo

Guedes

et al. 2018

Rev. Bras. Ciênc. Solo

View full text Add to dashboard Cite

In geostatistical modeling of soil chemical properties, one or more influential observations in a dataset may impair the construction of interpolation maps and their accuracy. An alternative to avoid the problem would be to use most robust models, based on distributions that have heavier tails. Therefore, this study proposes a spatial linear model based on the slash distribution (SSLM) in order to characterize the spatial variability of soybean yields as a function of soil chemical properties. The likelihood ratio statistic (LR) was applied to verify the significance of parameters associated with the model. We evaluated the sensitivity of the maximum likelihood estimators by means of local influence analysis for both the soybean response and the linear predictor. In the proposed model, we analyzed data gathered from a commercial grain production area (127.18 ha) located in the western part of the state of Paraná (Brazil). The results showed that the slash distribution allowed us to adjust the high kurtosis of the data set distribution and the LR test confirmed that the soil chemical properties of phosphorus, potassium, pH, and organic matter were significant for the SSLM. Diagnostic analysis indicated that the atypical value of the sample set was not influential in the parameter estimation process. Construction of the interpolation map based on the proposed model is not affected when considering the atypical and/or influential observations. Thus, SSLM becomes a robust alternative in the study of soybean yield variability as a function of soil chemical properties, making it possible to investigate the productive potential of the areas.

show abstract

Section: Introductionmentioning

confidence: 99%

Slash Spatial Linear Modeling: Soybean Yield Variability as a Function of Soil Chemical Properties

Fagundes

Uribe-Opazo

Guedes

et al. 2018

Rev. Bras. Ciênc. Solo

View full text Add to dashboard Cite

show abstract

“…Lange et al (1989) propose a reparametrization of the t-Student distribution from a transformation in the shape parameter v , allowing us to assume the existence of the second finite moment and thus a more direct comparison with the normal distribution. This reparametrization is justified by the importance that the spatial dependence modeling represents since the new shape parameter  is limited and this process allows estimating parameters by maximum likelihood (Nesi et al, 2013) and implementing the EM iterative algorithm (Dempster et al, 1977;Assumpção et al, 2014).…”

Section: Introductionmentioning

confidence: 99%

Spatial Variability of Soybean Yield Through a Reparameterized T-Student Model

et al. 2017

View full text Add to dashboard Cite

ABSTRACT:The t-Student distribution has been used to the spatial dependence modelling of soybean yield as an alternative to the normal distribution, being used for data with heavier tails or discrepant values. However, a usual Student t-distribution does not allow direct comparisons of geostatistical methods with a normal distribution. The aim of this study was to assess the soybean yield spatial variability through a reparameterized t-Student linear model, comparing the results with those of a Gaussian linear model. For parameter estimation, a complete maximum likelihood (CML) method was used through an expectation-maximization (EM) algorithm. The maps constructed with both reparameterized t-Student and normal distributions are dissimilar and present a kappa index (K) equivalent to 0.64. The reparameterized t-Student distribution is an alternative in studying data with discrepant values, showing the ability to decrease the influence of these points.

show abstract

“…Research on the spatial dependence structure of agricultural georeferenced variables, such as chemical and physical soil properties and crop yield, is an analysis tool which provides information to support a decision in favour of better management of production areas (BORSSOI et al, 2011;GREGO et al, 2011;URIBE-OPAZO et al, 2012;NESI et al, 2013;ASSUMPÇÃO et al, 2014;BERNARDI et al, 2014). This can be accomplished by means of geostatistical techniques that retrieve from a set of sample elements, information about the spatial variation of the phenomenon in the whole area through the construction of thematic maps of variability (DIGGLE & RIBEIRO JUNIOR, 2007).…”

Section: Introductionmentioning

confidence: 99%

Soybean yield maps using regular and optimized sample with different configurations by simulated annealing

et al. 2016

View full text Add to dashboard Cite

ABSTRACT:This study aimed to compare thematic maps of soybean yield for different sampling grids, using geostatistical methods (semivariance function and kriging). The analysis was performed with soybean yield data in t ha -1 in a commercial area with regular grids with distances between points of 25x25 m, 50x50 m, 75x75 m, 100x100 m, with 549, 188, 66 and 44 sampling points respectively; and data obtained by yield monitors. Optimized sampling schemes were also generated with the algorithm called Simulated Annealing, using maximization of the overall accuracy measure as a criterion for optimization. The results showed that sample size and sample density influenced the description of the spatial distribution of soybean yield. When the sample size was increased, there was an increased efficiency of thematic maps used to describe the spatial variability of soybean yield (higher values of accuracy indices and lower values for t he sum of squared estimation error). In addition, more accurate maps were obtained, especially considering the optimized sample configurations with 188 and 549 sample points.KEYWORDS: accuracy indices, optimization, sampling grids, spatial variability. MAPAS DA PRODUTIVIDADE DA SOJA USANDO CONFIGURAÇÕES AMOSTRAISREGULARES E OTIMIZADAS PELA TÊMPERA SIMULADA RESUMO: Neste trabalho, teve-se o objetivo de estudar e comparar mapas temáticos da produtividade da soja para diferentes grades amostrais, utilizando métodos geoestatísticos (função semivariância e krigagem). A análise foi realizada com dados da produtividade de soja em t ha -1 numa área comercial com grades regulares e distâncias entre pontos de 25x25 m, 50x50 m, 75x75 m e 100x100 m, com 549; 188; 66 e 44 pontos amostrais, respectivamente, e dados obtidos pelo monitor de colheita. Também foram gerados esquemas amostrais otimizados, com o algoritmo chamado Têmpera Simulada, usando a maximização da medida de acurácia exatidão global, como critério de otimização. Foi verificado que fatores como o tamanho da amostra e a densidade amostral influenciaram na descrição da distribuição espacial da produtividade da soja feita por meio dos mapas temáticos. Quando se aumentou o tamanho amostral, houve aumento da eficiência dos mapas temáticos quanto à descrição da variabilidade espacial da produtividade da soja (maiores valores dos índices de acurácia e os menores valores para a soma quadrada do erro de estimação). Além disso, os mapas mais acurados foram obtidos considerando as configurações amostrais otimizadas com 188 e 549 pontos amostrais.PALAVRAS-CHAVE: grades amostrais, índices de acurácia, otimização, variabilidade espacial.Soybean yield maps using regular and optimized sample with different configurations by simulated annealingEng.

show abstract

Verossimilhança na seleção de modelos para predição espacial

Cited by 5 publications

References 20 publications

Slash Spatial Linear Modeling: Soybean Yield Variability as a Function of Soil Chemical Properties

Slash Spatial Linear Modeling: Soybean Yield Variability as a Function of Soil Chemical Properties

Spatial Variability of Soybean Yield Through a Reparameterized T-Student Model

Soybean yield maps using regular and optimized sample with different configurations by simulated annealing

Contact Info

Product

Resources

About