[adegenet-forum] $li score in sPCA

[Jombart, Thibaut]

Hello, 

as in any multivariate analysis, the sign of the PCs is arbitrary. Only the distance between individuals on this PC has a meaning, i.e. if you have (using integers to make things simpler):
A = -1
B = 1
C = 3
D = 5

Then the difference between A and C is the same as between B and D.
Cheers
Thibaut
________________________________________
From: adegenet-forum-bounces at lists.r-forge.r-project.org [adegenet-forum-bounces at lists.r-forge.r-project.org] on behalf of Ting Pu [tingpu89 at gmail.com]
Sent: 20 October 2014 18:18
To: adegenet-forum at lists.r-forge.r-project.org
Subject: [adegenet-forum] $li score in sPCA

Hi all,

I was just wondering in sPCA, after I have selected the first positive principal component (which represents global structures), how should I interpret the positiveness and negativeness of the $li (entity scores)? Does a high positive $li mean its spatial correlation is stronger than a negative li? Please correct me if I am wrong.

Thank you for your time,

Ting
_______________________________________________
adegenet-forum mailing list
adegenet-forum at lists.r-forge.r-project.org
https://lists.r-forge.r-project.org/cgi-bin/mailman/listinfo/adegenet-forum