Abstract
Agronomists and breeders frequently collect yield data for a number of genotypes in a number of environments (site-years), resulting in a two-way data table. The Additive Main effects and Multiplicative Interaction (AMMI) model combines regular analysis of variance (ANOVA) for additive main effects with principal components analysis (PCA) for multiplicative structure within the interaction (that is, within the residual from ANOVA). AMMI is effective for (1) understanding genotype-environment interaction, (2) improving the accuracy of yield estimates, (3) increasing the probability of successfully selecting genotypes with the highest yields, (4) imputing missing data, and (5) increasing the flexibility and efficiency of experimental designs. Ultimately these advantages imply larger selection gains in breeding research and more reliable recommendations in agronomy research. AMMI is ordinarily the statistical method of choice when main effects and interaction are both important.
Keywords
AMMI, ANOVA
Creative Commons License
This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.
Recommended Citation
Gauch, Hugh G. Jr. and Zobel, Richard W.
(1989).
"USING INTERACTION IN TWO-WAY DATA TABLES,"
Conference on Applied Statistics in Agriculture.
https://doi.org/10.4148/2475-7772.1467
USING INTERACTION IN TWO-WAY DATA TABLES
Agronomists and breeders frequently collect yield data for a number of genotypes in a number of environments (site-years), resulting in a two-way data table. The Additive Main effects and Multiplicative Interaction (AMMI) model combines regular analysis of variance (ANOVA) for additive main effects with principal components analysis (PCA) for multiplicative structure within the interaction (that is, within the residual from ANOVA). AMMI is effective for (1) understanding genotype-environment interaction, (2) improving the accuracy of yield estimates, (3) increasing the probability of successfully selecting genotypes with the highest yields, (4) imputing missing data, and (5) increasing the flexibility and efficiency of experimental designs. Ultimately these advantages imply larger selection gains in breeding research and more reliable recommendations in agronomy research. AMMI is ordinarily the statistical method of choice when main effects and interaction are both important.
