SAS变量聚类过程.docVIP

SAS变量聚类过程 13个上市公司的财务状况指标 Num：股票代码 X1-流动比率 X1-速动比率 X1-总资产周转率 X1-存货周转率 X1-营运资本 X1-每股收益 X1-净利润增长率 X1-每股收益增长率 X1-主营业务毛利率 X1-主营业务利润率 X1-成本费用利润率 X1-净资产收益率 X1-总资产利润率 一、变量聚类过程代码 Proc varclus data=a maxclusters=3; Var x1-x13; Run; 二、输出结果 Observations 48 Proportion 0 Variables 13 Maxeigen 1 Cluster Variation Proportion Second Cluster Members Variation Explained Explained Eigenvalue 1 13 13 6.643831 0.5111 2.3926 Total variation explained = 6.643831 Proportion = 0.5111 Cluster 1 will be split because it has the largest second eigenvalue, 2.392611, which is greater than the MAXEIGEN=1 value. Clustering algorithm converged. Cluster Summary for 2 Clusters Cluster Variation Proportion Second Cluster Members Variation Explained Explained Eigenvalue ------------------------------------------------------------------------ 1 10 10 6.1972 0.6197 1.4838 2 3 3 2.67548 0.8918 0.2719 Total variation explained = 8.872681 Proportion = 0.6825 R-squared with 2 Clusters ------------------ Own Next 1-R**2 Variable Cluster Variable Cluster Closest Ratio Label ------------------------------------------------------------------ Cluster 1 X3 0.3876 0.0046 0.6152 X3 X4 0.3474 0.0003 0.6527 X4 X6 0.8843 0.0879 0.1269 X6