limix.stats.pca

limix.stats.pca(X, ncomp)

Principal component analysis.

Parameters:

• X (array_like) – Data.
• ncomp (int) – Number of components.

Returns:

• components (array_like): first components ordered by explained variance.
• explained_variance (array_like): explained variance.
• explained_variance_ratio (array_like): percentage of variance explained.

Return type:

dict

Examples

>>> from numpy import round
>>> from numpy.random import RandomState
>>> from limix.stats import pca
>>>
>>> X = RandomState(1).randn(4, 5)
>>> r = pca(X, ncomp=2)
>>> round(r['components'], 2)
array([[-0.75,  0.58, -0.08,  0.2 , -0.23],
       [ 0.49,  0.72,  0.02, -0.46, -0.16]])
>>> round(r['explained_variance'], 4) 
array([ 6.4466,  0.5145])
>>> round(r['explained_variance_ratio'], 4) 
array([ 0.9205,  0.0735])