实验报告9典型相关分析报告

《实验报告9典型相关分析报告》由会员上传分享，免费在线阅读，更多相关内容在行业资料-天天文库。

1、......实验九典型相关分析实验目的和要求能利用原始数据与相关矩阵、协主差矩阵作相关分析,能根据SAS输出结果选出满足要求的几个典型变量．实验要求:编写程序，结果分析．实验内容：4.8SAS实现dataexamp4_8(type=corr);input_name_$x1-x2y1-y2; _type_=’corr’;cards;x11.000.630.240.06x20.631.00-0.060.07y10.24-0.061.000.42y20.060.070.421.00;run;proccancorrdata

2、=examp4_8corr;varx1-x2;withy1-y2;run;TheSASSystem20:05Thursday,November18,20131TheCANCORRProcedureCorrelationsAmongtheOriginalVariables1、变量x1-x2的相关系数矩阵：CorrelationsAmongtheVARVariablesx1x2x11.00000.6300x20.63001.00002、变量y1-y2的相关系数矩阵：CorrelationsAmongtheWITHVar

3、iables学习好帮手......y1y2y11.00000.4200y20.42001.00003、变量x1-x2与y1-y2的相关系数矩阵：CorrelationsBetweentheVARVariablesandtheWITHVariablesy1y2x10.24000.0600x2-0.06000.0700变量间高度相关。TheSASSystem20:05Thursday,November18,20132TheCANCORRProcedure4典型相关分析的一般结果CanonicalCorrelationA

4、nalysisAdjustedApproximateSquaredCanonicalCanonicalStandardCanonicalCorrelationCorrelationErrorCorrelation典型相关系数校正的典型相关系数近似的标准误典型相关系数平方10.3971120.3969100.0084230.15769820.072889.0.0099470.0053135、检验各对典型变量是否显著相关TestofH0:ThecanonicalcorrelationsintheEigenvalueso

5、fInv(E)*Hcurrentrowandallthatfollowarezero=CanRsq/(1-CanRsq)LikelihoodApproximateEigenvalueDifferenceProportionCumulativeRatioFValueNumDFDenDFPr>F各对相关系相邻两特特征值占特征值占方差似然比值数特征值征值之差方差比例比例累计值10.18720.18190.97230.97230.83782737462.33419992<.000120.00530.02771.00000.

6、9946871253.4019997<.0001第一对典型变量贡献率97.23%。充分反映了两组变量的相互关系。学习好帮手......检验假设检验统计量，为第一、第二自由度．由检验结果可知，，．故两对典型变量显著相关．取两对进行分析即可．另外，从对典型变量进行分析求得特征值在方差占比例的累计值（贡献率）为0.9947也可看出，两对变量即可。以下输出用wilks’Lambda等四种方法对典型相关系数为零的假设检验6、求出典型变量及典型相关系数，并解释典型变量的系数和典型结构MultivariateStatistics

7、andFApproximationsS=2M=-0.5N=4997StatisticValueFValueNumDFDenDFPr>FWilks'Lambda0.83782737462.33419992<.0001Pillai'sTrace0.16301046443.56419994<.0001Hotelling-LawleyTrace0.19256330481.20411994<.0001Roy'sGreatestRoot0.18722205935.8329997<.0001NOTE:FStatisticforR

8、oy'sGreatestRootisanupperbound.NOTE:FStatisticforWilks'Lambdaisexact.TheSASSystem20:05Thursday,November18,20133TheCANCORRProcedureCanonicalCorrelationAnalysisRawCanonicalCoefficien