Canonical correlations in studying grain yield and protein content as affected by yield components: An ontogenetic approach.

Marcin Kozak

m.kozak@omega.sggw.waw.pl
Department of Biometry Warsaw Agricultural University, Nowoursynowska 159, 02-776 Warsaw (Poland)

Dariusz Gozdowski

Department of Biometry Warsaw Agricultural University, Nowoursynowska 159, 02-776 Warsaw (Poland)

Shakhawat Hossain

Department of Mathematics & Statistics, University of Windsor, Canada (Canada)

S. Ejaz Ahmed

Department of Mathematics & Statistics, University of Windsor, Canada (Canada)

Zbigniew Laudański

Plant Breeding and Acclimatization Institute, Radzików, 05-870 Błonie, Poland (Poland)

Zdzisław Wyszyński

Department of Agronomy, Warsaw Agricultural University, Nowoursynowska 159, 02-776 Warsaw (Poland)

Abstract

In the paper, we have proposed an approach to studying the relationship between two sets of variables, when one of the sets comprises plant traits that develop sequentially. The approach joins two statistical methods, namely canonical correlations and sequential yield analysis (Mądry et al., 2005, J. New Seeds 7(1), pp. 85-107). Using the approach, grain yield and protein content in grain of two spring barley cultivars—Rasbet (with hulled grains) and Rastik (with hulless grains)—were studied as the effect of yield components that develop sequentially (number of spikes per m², number of kernels per spike, and kernel weight)...

Keywords:

multivariate methods, sequential yield analysis, spring barley