Hello, I have been employing GRASS 6.2 to do Principal Components Analysis (PCA) on a series of images using the i.pca command. The output of eigenvectors looks like the following:
( 0.55 0.78 0.28 ) ( -0.17 -0.22 0.96 ) ( 0.82 -0.58 0.01 ) However, lacking training in matrix algebra, I am unsure about how to generate eigen values and percent variance explained by the different principal components from the eigen vectors. The GRASS 6.4 documentation for i.pca indicates that this alternative information is included in the new output (see the documentation example below): r.info -h spot_pca.1 Eigen values, (vectors), and [percent importance]: PC1 1170.12 ( -0.63 -0.65 -0.43 ) [ 88.07% ] PC2 152.49 ( 0.23 0.37 -0.90 ) [ 11.48% ] PC3 6.01 ( 0.75 -0.66 -0.08 ) [ 0.45% ] After installing GRASS 6.4, however, the output claims to be eigen values, but looks more like eigen vectors as for GRASS 6.2: ( 0.37 0.46 0.81 ) ( -0.50 -0.63 0.59 ) ( 0.78 -0.63 -0.01 ) I would appreciate any help on interpreting these values, or knowing why the output of i.pca in GRASS 6.4 doesn't match what is in the documentation. Ultimately, I am only trying to get at the percent variance for each principal component so that I can proceed with supervised and unsupervised classification of my images. Thanks much, William _______________________________________________ grass-user mailing list [email protected] http://lists.osgeo.org/mailman/listinfo/grass-user
