Soybean yield modeling using bootstrap methods for small samples

Soybean yield modeling using bootstrap methods for small samples

Gustavo H. Dalposso, Miguel A. Uribe-Opazo and Jerry A. Johann

http://dx.doi.org/10.5424/sjar/2016143-8635

Abstract: One of the problems that occur when working with regression models is regarding the sample size; once the statistical methodsused in inferential analyzes are asymptotic if the sample is small the analysis may be compromised because the estimates will bebiased. An alternative is to use the bootstrap methodology, which in its non-parametric version does not need to guess or know theprobability distribution that generated the original sample. In this work we used a set of soybean yield data and physical andchemical soil properties formed with fewer samples to determine a multiple linear regression model. Bootstrap methods were usedfor variable selection, identification of influential points and for determination of confidence intervals of the model parameters. Theresults showed that the bootstrap methods enabled us to select the physical and chemical soil properties, which were significant inthe construction of the soybean yield regression model, construct the confidence intervals of the parameters and identify the pointsthat had great influence on the estimated parameters.Additional

key words: multiple linear regression; model selection; bootstrap global influence diagnosis; bootstrap confidence intervals.

 

Publish modules to the "offcanvas" position.

Nós usamos cookies
Nós usamos cookies no nosso site. Alguns deles são essenciais para o funcionamento do site, enquanto que outros ajudam a melhorar o site e a experiência do usuário. Você pode decidir se quer permiti-los ou não. Tenha em mente que, em caso de rejeição, você não será capaz de usar todas as funcionalidades do site.