Cluster Analysis: 2014 Edition (Statistical Associates Blue Book Series 24)

Cluster Analysis: 2014 Edition (Statistical Associates Blue Book Series 24)

CLUSTER ANALYSIS Overview<br /><br />An illustrated tutorial and introduction to cluster analysis using SPSS, SAS, SAS Enterprise Miner, and Stata for examples. Suitable for introductory graduate-level study.<br /><br />The 2014 edition is a major update to the 2012 edition. Among the new features are these:<br /><br /> Was 89 pages, now book length (207 pages total)<br /> Had 58 figures, now has over 170 illustrations<br /> Now covers Stata as well as SPSS and SAS<br /> Totally revised sections on hierarchical, k-means, and two-step clustering<br /> New coverage of nearest neighbor analysis<br /> New coverage of oblique principal components cluster analysis<br /> New coverage of nonparametric density cluster analysis<br /> New coverage of Kohonen self-organizing map (SOM) clustering<br /> Links to all datasets used in the text. <br /><br />Below is the unformatted partial table of contents.<br /><br />Table of Contents<br />CLUSTER ANALYSIS 1<br />Overview10<br />Data examples in this volume10<br />Key Concepts and Terms12<br />Terminology12<br />Distances (proximities)12<br />Cluster formation12<br />Cluster validity12<br />Types of cluster analysis14<br />Types of cluster analysis by software package14<br />Disjoint clustering15<br />Hierarchical clustering15<br />Overlapping clustering16<br />Fuzzy clustering16<br />Hierarchical cluster analysis in SPSS16<br />SPSS Input for hierarchical clustering16<br />Example16<br />The main "Hierarchical Cluster Analysis" dialog17<br />Statistics button18<br />Plots button19<br />Methods button20<br />SPSS output for hierarchical cluster analysis21<br />Proximity table21<br />Cluster membership table22<br />Agglomeration Schedule22<br />Dendogram24<br />Icicle plots27<br />Summary measures28<br />Hierarchical cluster analysis in SAS29<br />SAS input for hierarchical cluster analysis29<br />Example29<br />Data setup29<br />SAS syntax30<br />SAS output for hierarchical cluster analysis31<br />Simple statistics table31<br />Eigenvalues of the covariance matrix table31<br />Root mean square coefficients32<br />Cluster history table33<br />Dendogram34<br />Icicle Plots36<br />Cluster membership table36<br />Saving data to file37<br />Hierarchical cluster analysis in Stata38<br />Stata input for hierarchical cluster analysis38<br />Stata output for hierarchical cluster analysis40<br />Agglomeration coefficients40<br />Dendogram41<br />Saving cluster membership values42<br />Cluster membership table43<br />K-means cluster analysis44<br />Overview44<br />Example45<br />K-means cluster analysis in SPSS45<br />SPSS input45<br />Main K-means dialog45<br />The Iterate button47<br />The Save button48<br />The Options button49<br />SPSS Output for K-Means cluster analysis50<br />The Anova table50<br />Number of cases in each cluster51<br />Getting different clusters52<br />Cluster membership table52<br />K-Means cluster analysis in SAS53<br />Overview53<br />Example54<br />SAS input for k-means cluster analysis54<br />SAS output for k-means cluster analysis55<br />The "Statistics for Variables" table55<br />Criteria for determining k57<br />The "Cluster Summary" table60<br />Cluster membership and distance values61<br />Crosstabulation tables61<br />Cluster separation plots62<br />K-Means cluster analysis in Stata64<br />Example64<br />Stata input for k-means cluster analysis64<br />The main kmeans clustering command64<br />Obtaining descriptive statistics65<br />Obtaining distance information65<br />Obtaining cluster separation plots65<br />Comparing kmeans and kmedian solutions66<br />Stata output for k-means cluster analysis66<br />Cluster membership assignments66<br />Descriptive statistics67<br />Distance coefficients69<br />Cluster separation plots70<br />Comparing kmeans and kmedians solutions71<br />Two-step cluster analysis in SPSS72<br />Overview72<br />Cluster feature tree (CF tree)73<br />Proximity73<br />Example74<br />SPSS input for two-step clustering74<br />The main two-step clustering dialog74<br />Options button dialog75<br />Output button dialog78<br />SPSS output for two-step clustering79<br />Autoclustering table79<br />Cluster distribution table81<br />Centroids (cluster profiles) table81<br />Model summary82<br />The "Cluster Quality" graph82<br />The "Cluster Sizes" pie chart82<br />The "Predictor Importance" chart83