ISHS Acta Horticulturae 313: III International Symposium on Computer Modelling in Fruit Research and Orchard Management
CANONICAL DISCRIMINANT ANALYSIS IN KIWIFRUIT ROOTSTOCK RESEARCH |
| Authors: | J.G. Cruz-Castillo, B.R. MacKay, G.S. Lawes, D.J. Woolley |
| Abstract:
Canonical discriminant analysis (CDA) is a multivariate statistical technique that maximises the correlation between two sets of variables, generating linear combinations that can discriminate between groups of individuals. CDA has not been used extensively in horticultural research. It has applications when is desirable to find differences between groups of individuals or treatments, for example cultivars (Yourstone and Wallace, 1990) growth regulators, levels of irrigation, tree training systems, and where it is desired to understand the relationship between the several variables measured on those groups. In a kiwifruit rootstock trial, CDA was used to separate vines on nine different rootstocks on the basis of yield, fruit size, number of fruits removed by thinning, percentage of flowers in a compound dichasium ('triples'), and number of flowers per shoot. Plants on two rootstocks achieved a higher field performance than vines on the remaining rootstocks, an effect which was associated with a heavy yield of large sized fruit. | |
Download Adobe Acrobat Reader (free software to read PDF files) | |
