HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Full Text Free Full Text (PDF) Free Alert me when this article is cited Alert me if a correction is posted Services Similar articles in this journal Similar articles in ISI Web of Science Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via ISI Web of Science (5) Citing Articles via Google Scholar Google Scholar Articles by Piepho, H.-P. Articles by Möhring, J. Search for Related Content PubMed Articles by Piepho, H.-P. Articles by Möhring, J. Agricola Articles by Piepho, H.-P. Articles by Möhring, J. Related Collections Biometrics Experiment Design Statistics

CROP BREEDING, GENETICS & CYTOLOGY

Selection in Cultivar Trials—Is It Ignorable?

Bioinformatics Unit, Institut für Pflanzenbau und Grünland, Universität Hohenheim, 70599 Stuttgart, Germany

^* Corresponding author (piepho{at}uni-hohenheim.de)

Crop cultivar registration requires multienvironment trials for assessing the value for cultivation and use (VCU). The series of trials usually extends across 3 yr, with some cultivars being discarded each year. Selection gives rise to a missing-data or drop-out pattern that is not completely random. The present paper studies the effect of drop-out on the validity of mixed model procedures such as REML and BLUP. It is shown on the basis of the pertinent statistical theory and simulations that selection is ignorable providing that all data used in the selection process are included in the analysis. Simulations show that REML is preferable to ML and BLUP is preferable to BLUE. It is suggested that cultivar registration authorities can benefit from multivariate mixed model analyses comprising all traits on which selection is based.

Abbreviations: BLUE, best linear unbiased estimation • BLUP, best linear unbiased prediction • MAR, missing at random • MCAR, missing completely at random • ML, maximum likelihood • MNAR, missing not at random • MSE, mean squared error • REML, restricted (residual) maximum likelihood • VCU, value for cultivation and use