Editing Linear discriminant analysis (section)

==Canonical discriminant analysis for ''k'' classes==
Canonical discriminant analysis (CDA) finds axes (''k''&nbsp;−&nbsp;1 [[canonical coordinates]], ''k'' being the number of classes) that best separate the categories. These linear functions are uncorrelated and define, in effect, an optimal ''k''&nbsp;−&nbsp;1 space through the ''n''-dimensional cloud of data that best separates (the projections in that space of) the ''k'' groups. See “[[#Multiclass LDA|Multiclass LDA]]” for details below.

Because LDA uses canonical variates, it was initially often referred as the "method of canonical variates"<ref>{{cite book|last=Nabney|first=Ian|title=Netlab: Algorithms for Pattern Recognition|year=2002|page=274|ISBN=1-85233-440-1}}</ref> or canonical variates analysis (CVA).<ref>{{cite book|year=2023|title=Statistical Computing for Biologists|last=Magwene|first=Paul|chapter=Chapter 14: Canonical Variates Analysis|url=https://bio723-class.github.io/Bio723-book/canonical-variates-analysis.html}}</ref>