Merakit Sifat Ketegaran Terhadap Ketaknormalan Data dan Pengamatan Pencilan Pada Model AMMI
Abstract
AMMI (Additive Main Effect Multiplicative Interaction) model for interactions in two-way table provide the major mean for studying stability and adaptability through genotype × environment interaction (GEI), which modeled by full interaction model. Eligibility of AMMI model depends on that assumption of normally independent distributed error with a constant variance. Nowadays, AMMI models have been developed for any condition of MET data with some violence of the normality and homegeneity assumption. We can mention in this class of medelling as M-AMMI for mixed AMMI models and G-AMMI for generalized AMMI models. The G-AMMI was handling non-normality i.e categorical response variables using an algorithm of alternating regression. Modeling count data in study of incidence on a plant for example, the appropriateness of AMMI model here is being doubtful. GAMMI log-link model will be applied to the Poisson data distribution. GAMMI log-link models give us good information of the interaction by its log-odd ratio. While in handling the non-homogeneity in mix-models sense, one may use a model called factor analytic multiplicative. The development of AMMI models is also to handle any outlier that might be found coincides with non-homogeneity condition of the data. In this paper, we will present both of handling nonnormality and outling observation in AMMI model by using an algorithm of alternating regression.