CMStatistics 2018: Start Registration
View Submission - CMStatistics
Title: Robust estimation in plant breeding: Evaluation using simulation and empirical data Authors:  Vanda Lourenco - Faculty of Sciences and Technology - New University of Lisbon (Portugal) [presenting]
Hans-Peter Piepho - Universitaet Hohenheim (Germany)
Joseph O. Ogutu - Bioinformatics Unit - Institute of Crop Science - University of Hohenheim (Germany)
Abstract: Genomic prediction (GP) is used to determine the best genotypes for selection in plant breeding. Accurate estimation of predictive accuracy (PA) that measures the effectiveness of GP is thus of paramount importance for GP. Regression models are the models of choice for analyzing field data in plant breeding. However, when their underlying assumptions are violated, models that use the classical likelihood typically perform poorly, often resulting in biased parameter estimates. In plant genetics such biases usually result in inaccurate estimates of heritability (H) and predictive accuracy, and hence compromise the predictive performance of GP. Since phenotypic data are susceptible to contamination, improving the methods for estimating heritability and predictive accuracy can enhance the performance of GP. Robust statistical methods provide an intuitively appealing and a theoretically well justified framework for overcoming some of the drawbacks of classical regression. We introduce and evaluate the performance of a robust approach to two recently proposed methods from the literature for estimating heritability and predictive accuracy of GP against the classical approach through simulation under several plausible scenarios of data contamination. An example application to a rye dataset is presented and used to empirically assess the adequacy and usefulness of the robust approach.