Optional. Type or paste the categories/species of each observation. For color-coding the plot.

#variables (rows):
#observations per variable (columns):

Variables' statistics:
variable# mean s

Covariance matrix: symmetric

Data centered at origin and scaled by SD, i.e. standardized:

Their statistics: should be 0 and 1
variable# mean s

Their Covariance matrix K (=correlation matrix):

Their eigenvalues λ_i, eigenvectors e_i:          Magnitudes |e_i|
of a square symmetric matrix: all eigenvectors are orthogonal, all eigenvalues are [non-negative?] real.
DAMN: the math.js eigs() gives normed eigenvectors.... OK for PCs etc. that follows.

λ1		e1		|e1|
λ2		e2		|e2|
λ3		e3		|e3|
λ4		e4		|e4|
λ5		e5		|e5|

Loadings/weights: each PC row's α's i.e. NORMED
eigenvalues λ_i, eigenvectors e_i          Magnitudes |e_i| (=1)

λ1		e1		|e1|
λ2		e2		|e2|
λ3		e3		|e3|
λ4		e4		|e4|
λ5		e5		|e5|

standardized Covariance matrix times each normed eigenvector equals normed eigenvalue times normed eigenvector: Ke_i=λ_ie_i
eigenvectors are only stretched by matrix (not any rotation)

Sum of eigenvalues = sum of its matrix K's diagonal (its trace)
Product of eigenvalues = det(matrix K)

Primary components Scores of standardized data:

PC1
PC2
PC3
PC4
PC5

Their statistics:
# mean s Var
Total VARiation:
Screes:

PC1+PC2 % of total VARiation:

X min: X max: Y min: Y max: