12.07.2015 Views

Jolliffe I. Principal Component Analysis (2ed., Springer, 2002)(518s)

Jolliffe I. Principal Component Analysis (2ed., Springer, 2002)(518s)

Jolliffe I. Principal Component Analysis (2ed., Springer, 2002)(518s)

SHOW MORE
SHOW LESS
  • No tags were found...

You also want an ePaper? Increase the reach of your titles

YUMPU automatically turns print PDFs into web optimized ePapers that Google loves.

Index 473variance ellipsoids,S-estimatorsrotationof factors 116, 153–156, 159,162–165of functional PCs 316, 327of PCs 43, 74, 151, 154, 162,163, 165, 166, 182, 185, 188,191, 213, 238, 248, 269–279,291, 295, 297, 298, 370, 396,407of subsets of PCs with similarvariances 276, 277rotation/switching of PCs 259to simple components 285, 291to simple structure 153, 154,182, 185, 270, 271, 276, 277,369see also oblique factors,orthogonal factors,quartimin rotation, varimaxrotationrounded PC coefficients 40, 42–44,67, 259, 263, 292, 293see also discrete PC coefficients,simplified PC coefficientsrow-centered data 89rules for selecting PCs 54, 111–137,159, 162, 217ad hoc rules 112–118, 130–135,136, 138, 147, 149, 238based on cross-validation 112,120–127, 130–132, 135–137based on cumulative variances ofPCs 112–114, 117, 130–136,138, 147, 147based on gaps betweeneigenvalues 126, 127, 129,133based on individual variances ofPCs 114–118, 123, 130–136,138, 147, 149, 238based on partial correlation 127from atmospheric science 112,116, 118, 127–130, 132–136statistically based rules 112,118–137see also broken stick model,equal variance PCs, howmany PCs, Kaiser’s rule,log eigenvalue diagram,parallel analysis, screegraph, selection of a subsetof PCsRV-coefficient 38, 143–145, 147,252S-estimators 267S-mode analysis 308, 398sample sizeseffective/equivalent 129, 148,299large see large data setsmoderate 249, 252small 65, 68, 148, 235, 257smaller than number of variables90, 148, 207, 413sample surveys 49, 328, 335, 336,353, 353stratified survey design 336, 353sampling variationPC coefficients 65PC variances 115, 123scale dependence of covariancematrices 24, 26see also invariance (scaleinvariance)scaling or ordination techniques85–90, 102, 106, 107, 200classical scaling 85dual scaling 103, 343non-metric multidimensionalscaling 86, 372reciprocal averaging 103, 343see also principal co-ordinateanalysisscatter, definitions of 395scores for PCs, see PC scoresSCoT (simplified componenttechnique) 278–279, 287,

Hooray! Your file is uploaded and ready to be published.

Saved successfully!

Ooh no, something went wrong!