Sparse Principal Component Analysis via Fractional Function Regularity

Table 3

Loadings and variance of numerical results of PCA, SPCA, and FP-SPCA methods in Case 2, where the FP-SPCA method works better than SPCA in obtaining the sparse loadings. (%) denotes the adjusted variance.

	PCA			SPCA		FP-SPCA
	PC1	PC2	PC3	PC1	PC2	PC1	PC2
	−0.3472	−0.3503	0.0819	−0.3565	−0.3466	−0.5	0
	−0.3472	−0.3503	0.0819	−0.3565	−0.3466	−0.5	0
	−0.3472	−0.3503	0.0819	−0.3565	−0.3466	−0.5	0
	−0.3472	−0.3503	0.0819	−0.3565	−0.3466	−0.5	0
	−0.3444	0.3566	0.0651	−0.3505	0.3604	0	0.5
	−0.3444	0.3566	0.0651	−0.3505	0.3604	0	0.5
	−0.3444	0.3566	0.0651	−0.3505	0.3604	0	0.5
	−0.3444	0.3566	0.0651	−0.3505	0.3604	0	0.5
	−0.1472	−0.0159	−0.6914	0	0	0	0
	−0.1472	−0.0159	−0.6914	0	0	0	0
(%)	49.23	32.10	18.67	44.61	29.88	40.00	38.44