Robustness of the Joint Regression Analysis

Abstract

Joint Regression Analysis is shown to be extremely robust to missing observations. Thus, using a series of "α-designs" of winter rye cultivars, it was shown that with up to 40% of missing observations the cultivars to be selected would be the same. In this study we considered missing observations incidences varying from 5% to 75% with 5% differences between them. For each incidence the positions of missing observations were randomly generated in triplicate.