Abstract

In crop science, genotype-by-environment interaction is oftenexplored using the "genotype main effects and genotype-by-environmentinteraction effects" (GGE) model. Using this model, a singularvalue decomposition is performed on the matrix of residuals from a fit ofa linear model with main effects of environments. Provided that errorsare independent, normally distributed and homoscedastic, the significanceof the multiplicative terms of the GGE model can be tested usingresampling methods. The GGE method is closely related to principalcomponent analysis (PCA). The present paper describes i) the GGEmodel, ii) the simple parametric bootstrap method for testing multiplicativegenotype-by-environment interaction terms, and iii) how thisresampling method can also be used for testing principal components inPCA

Published in

Acta et Commentationes Universitatis Tartuensis de Mathematica
2015, volume: 19, number: 1, pages: 27-33
Publisher: Faculty of Mathematics and Computer Science, University of Tartu

Authors' information