《经济分析方法与手段》实验分析报告

 附件一、

1 陶瓷产量与城镇住宅建筑面积的相关分析

散点图

2 与新增医疗卫生机构面积的关系

3 与新增办公楼面积的关系

附件二、

说明：如果实验分析需要用到SPSS的图表，依次在附件中列出，并在“实验结果及分析”中相应说明。

 

第二篇：实验报告9 典型相关分析

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：4.9

方法一：SAS

data examp4_9;

input x1-x2 y1-y2;

cards;

191 155 179 145

195 149 201 152

181 148 185 149

183 153 188 149

176 144 171 142

208 157 192 152

189 150 190 149

197 159 189 152

188 152 197 159

192 150 187 151

179 158 186 148

183 147 174 147

174 150 185 152

190 159 195 157

188 151 187 158

163 137 161 130

195 155 183 158

186 153 173 148

181 145 182 146

175 140 165 137

192 154 185 152

174 143 178 147

176 139 176 143

197 167 200 158

190163187150;

run;

proccancorr data=examp4_9 corr;

var x1-x2;                         

with y1-y2;                         

run;

                                            The SAS System          16:48 Sunday, October 31, 2012   1

                                        The CANCORR Procedure

                              Correlations Among the Original Variables

                       Correlations Among the VAR Variables（变量x1-x2的相关系数矩阵）

                                                  x1                x2

                                x1            1.0000            0.7504

                                x2            0.7504            1.0000

                      Correlations Among the WITH Variables（变量y1-y2的相关系数矩阵）

                                                  y1                y2

                                y1            1.0000            0.8397

                                y2            0.8397            1.0000

                     Correlations Between the VAR Variables and the WITH Variables

变量x1-x3与y1-y3的相关系数矩阵

                                                  y1                y2

                                x1            0.7092            0.7050

                                x2            0.7140            0.7440

                                            The SAS System          16:48 Sunday, October 31, 2012   2

                                        The CANCORR Procedure

                                    Canonical Correlation Analysis

                                             Adjusted    Approximate        Squared

                             Canonical      Canonical       Standard      Canonical

                           Correlation    Correlation          Error    Correlation

                         1    0.801420       0.788167       0.074591       0.642274

                         2    0.084522        .             0.207025       0.007144

                                                         Test of H0: The canonical correlations in the

                      Eigenvalues of Inv(E)*H              current row and all that follow are zero

                        = CanRsq/(1-CanRsq)

                                                        Likelihood Approximate

       Eigenvalue Difference Proportion Cumulative      Ratio     F Value Num DF Den DF Pr > F

   1     1.7954     1.7882     0.9960     0.9960 0.35517013        6.78      4     40 0.0003

   2     0.0072                0.0040     1.0000 0.99285597        0.15      1     21 0.7014

检验假设 

检验统计量，为第一、第二自由度．由检验结果可知，，．故只有第一对典型变量显著相关．取第一对进行分析即可．

                            Multivariate Statistics and F Approximations

                                        S=2    M=-0.5    N=9

           Statistic                        Value    F Value    Num DF    Den DF    Pr > F

           Wilks' Lambda               0.35517013       6.78         4        40    0.0003

           Pillai's Trace              0.64941830       5.05         4        42    0.0021

           Hotelling-Lawley Trace      1.80263313       8.88         4        23    0.0002

           Roy's Greatest Root         1.79543769      18.85         2        21    <.0001

                    NOTE: F Statistic for Roy's Greatest Root is an upper bound.

                            NOTE: F Statistic for Wilks' Lambda is exact.

                                            The SAS System          16:48 Sunday, October 31, 2012   3

                                        The CANCORR Procedure

                                    Canonical Correlation Analysis

                           Raw Canonical Coefficients for the VAR Variables

                                                  V1                V2

                                x1      0.0474933802      -0.144760311

                                x2      0.0840477791      0.1959848929

                           Raw Canonical Coefficients for the WITH Variables

                                                  W1                W2

                                y1      0.0448783575      -0.174269545

                                y2      0.0850389917      0.2549589923

数据未标准化结果，即利用协方差矩阵分析的结果

 

                                            The SAS System          16:48 Sunday, October 31, 2012   4

                                        The CANCORR Procedure

                                    Canonical Correlation Analysis

                       Standardized Canonical Coefficients for the VAR Variables

                                                  V1            V2

                                    x1        0.4716       -1.4375

                                    x2        0.5963        1.3904

                      Standardized Canonical Coefficients for the WITH Variables

                                                  W1            W2

                                    y1        0.4593       -1.7835

                                    y2        0.5827        1.7471

,.

第一对典型变量

 

 

                                            The SAS System          16:48 Sunday, October 31, 2012   5

                                        The CANCORR Procedure

                                         Canonical Structure

                 Correlations Between the VAR Variables and Their Canonical Variables

                                                  V1            V2

                                    x1        0.9191       -0.3941

                                    x2        0.9502        0.3117

                 Correlations Between the WITH Variables and Their Canonical Variables

                                                  W1            W2

                                    y1        0.9486       -0.3164

                                    y2        0.9684        0.2494

       Correlations Between the VAR Variables and the Canonical Variables of the WITH Variables

                                                  W1            W2

                                    x1        0.7365       -0.0333

                                    x2        0.7615        0.0263

       Correlations Between the WITH Variables and the Canonical Variables of the VAR Variables

                                                  V1            V2

                                    y1        0.7602       -0.0267

                                    y2        0.7761        0.0211

原变量和第一对变量相关程度高，后一组提取的信息很少，与典型对系数一致。

方法二：MATLAB

>> a=[data];

[n,m]=size(a);

b=a./(ones(n,1)*std(a));

R=cov(b);

X=b(:,1:2);

Y=b(:,3:4);

[A,B,r,U,V,ststs]=canoncorr(X,Y)

A =

    0.5522   -1.3664

    0.5215    1.3784

B =

    0.5044   -1.7686

    0.5383    1.7586

r =

    0.7885    0.0537

U =

    0.5731   -0.0137

    0.3750   -1.6953

   -0.4877    0.0774

   -0.0209    0.7322

   -1.0535    0.0294

    1.6762   -2.0193

    0.1063   -0.6685

    1.1955   -0.1057

    0.1912   -0.1546

    0.2760   -1.0884

    0.1065    2.2268

   -0.4453   -0.3895

   -0.7422    1.4311

    0.7995    0.8741

    0.1205   -0.3416

   -2.2840    0.5404

    0.7994   -0.5736

    0.1488    0.3123

   -0.6999   -0.4835

   -1.3930   -0.5784

    0.5590   -0.3406

   -1.2373    0.1224

   -1.4072   -0.9053

    1.7614    1.3899

    1.0825    1.6219

V =

   -0.5833   -0.2587

    1.0836   -2.2993

    0.0390   -0.2672

    0.1898   -0.7957

   -1.2259    0.3643

    0.6314   -0.7140

    0.2902   -1.1480

    0.4807   -0.1856

    1.4442    0.2398

    0.3000   -0.0954

    0.0090   -0.7055

   -0.6741    1.1462

    0.2797    0.5190

    1.1832    0.0680

    0.8615    1.7392

   -2.6910   -1.0193

    0.6605    2.4438

   -0.6441    1.5845

   -0.3524   -0.5250

   -1.9285    0.1107

    0.2797    0.5190

   -0.4731    0.4416

   -0.8945   -0.2544

    1.5147   -0.5507

    0.2197   -0.3574

ststs =

     Wilks: [0.3772 0.9971]

       df1: [4 1]

       df2: [42 22]

         F: [6.5972 0.0637]

        pF: [3.2565e-004 0.8031]

     chisq: [20.9642 0.0639]

    pChisq: [3.2189e-004 0.8004]

       dfe: [4 1]

         p: [3.2189e-004 0.8004]