Premium
Analysing Your Multivariate Data as a Pictorial: A Case for Applying Biplot Methodology?
Author(s) -
Roux Niël J.,
Gardner Sugnet
Publication year - 2005
Publication title -
international statistical review
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.051
H-Index - 54
eISSN - 1751-5823
pISSN - 0306-7734
DOI - 10.1111/j.1751-5823.2005.tb00154.x
Subject(s) - biplot , multivariate statistics , multivariate analysis , mathematics , computer science , statistics , biology , biochemistry , gene , genotype
Summary This paper illustrates the versatility of biplot methodology when analysing multivariate data from diverse disciplines. The modern approach of Gower & Hand (1996) whereby biplots are regarded as multivariate analogues of ordinary scatter plots is utilised for extending biplot methodology introducing several novel applications. Focus is on biplot applications where the merits of principal component biplots and canonical variate analysis biplots are illustrated with data sets from higher education, the manufacturing industry, the mining industry, agriculture, finance and archaeology. It is shown how to equip biplots with quality regions, classification regions and acceptance regions; how a ‐bags superimposed on biplots provide a quantification of the multidimensional overlap of classes as well as enable biplots to be used with large data sets; how to use biplots for exploring multi‐dimensional reality and in sophisticated classification procedures.