7.2 Computer Science - Pattern Recognition

(These are most conveniently done with access to JMP (SAS, Inc.). Indeed Q.7.2 and 7.3 use datasets in the .JMP format (which can be downloaded from http://extras.springer.com ).

THIS is the books DL content - http://extras.springer.com/2013/978-1-4614-5322-2

7.2 - The data in socioeconomic.jmp consists of five socioeconomic variables/ features for 12 census tracts in the LA Metropolitan area. (a) Use the Multivariate platform to produce a scatterplot matrix of all five features. (b) Conduct a principal component analysis (on the correlations) of all five features. Considering the eigenvectors, which are the most useful features? Considering the eigenvalues, how many principal components would you use in subsequent analysis?