Output from mdsauto4.sas

Source
0 Graphs

Similarities of Automobiles 1

                 Multidimensional Scaling:  Data=WORK.DISSIM
                Shape=TRIANGLE Condition=MATRIX Level=ORDINAL
                  Coef=IDENTITY Dimension=2 Formula=1 Fit=1
        Mconverge=0.01 Gconverge=0.01 Maxiter=100 Over=2 Ridge=0.0001

                                                      Convergence Measures
                     Badness-of-Fit     Change in  --------------------------
Iteration  Type           Criterion     Criterion      Monotone      Gradient
-----------------------------------------------------------------------------
      0    Initial         0.223978             .             .             .
      1    Monotone        0.141124      0.082854      0.124573      0.550792
      2    Gau-New         0.119615      0.021509             .             .
      3    Monotone        0.113074      0.006541      0.034149      0.223075
      4    Gau-New         0.112929      0.000145             .             .
      5    Monotone        0.109139      0.003790      0.025829      0.177671
      6    Gau-New         0.108790      0.000348             .             .
      7    Monotone        0.106915      0.001876      0.020202      0.133907
      8    Gau-New         0.106766      0.000149             .             .
      9    Monotone        0.105485      0.001281      0.016298      0.109037
     10    Gau-New         0.105171      0.000314             .             .
     11    Monotone        0.104300      0.000871      0.013571      0.111673
     12    Gau-New         0.103765      0.000535             .             .
     13    Monotone        0.102633      0.001132      0.014273      0.111601
     14    Gau-New         0.102272      0.000361             .             .
     15    Monotone        0.101718      0.000554      0.010405      0.094706
     16    Gau-New         0.101653   0.000065102             .             .
     17    Monotone        0.101426      0.000227      0.006813      0.090674
     18    Gau-New         0.101096      0.000330             .      0.023329
     19    Gau-New         0.101076   0.000019704             .      0.006733

                     Convergence criteria are satisfied.

Similarities of Automobiles 2

              Plot of DIM2*DIM1$NAME.  Symbol points to label.

           |
       1.5 +
           |
           |
           |
           |                                   > Mustang
           |                    > 300ZX
           |                          > Corvette
       1.0 +
           |               > Porsche
           |
           |
           |
           |
           |
       0.5 +
           |
    D      |
    i      |
    m      |
    e      |                                               Escort <
    n      |
    s  0.0 +
    i      |
    o      |
    n      |
           |                                               Chevette <
    2      |
           |
      -0.5 +
           |             > Jaguar                            > Renault
           |
           |
           |
           |
           |
      -1.0 +             Cadillac <    > Lincoln
           |            > Mercedes
           |
           |
           |
           |
           |
      -1.5 +
           |
           ---+-------------+-------------+-------------+-------------+--
             -2            -1             0             1             2

                                     Dimension 1


Similarities of Automobiles 3

                      E         U                                  R    M
                 S    X         N         C                   E    E    A
                 O    C    S    C         O    S              L    L    S
                 P    I    T    O         M    P    F    S    E    I    C
     M           H    T    R    N    B    P    O    R    W    G    A    U
     A           I    I    O    V    O    L    R    E    I    A    B    L
     K           S    N    N    E    L    E    T    S    F    N    L    I
     E           T    G    G    N    D    X    Y    H    T    T    E    N
     MUSTANG   0.3  1.5  1.7  1.2  0.5  1.6  0.9  0.5  2.6  1.5  0.8  1.2
     CADILLAC  2.0 -1.0  1.2  0.5 -0.2  0.3 -1.8 -2.0  1.5  0.5  0.4 -0.5
     LINCOLN   2.2 -0.8  1.5  1.0  0.3  0.8 -2.6 -1.5  1.8  0.8  1.3  0.4
     FORD     -1.0 -2.0 -1.4 -1.4 -1.8 -2.4 -0.3  0.3 -0.4 -2.0  0.2 -0.8
     CORVETTE  1.2  2.0  1.8  1.3  2.2 -0.1  1.5  0.7  2.2 -0.8  0.9  1.3
     CHEVETTE -2.7 -1.8 -1.6 -2.0 -2.0 -2.5  0.2 -0.5 -2.5 -2.8  0.0 -1.2
     NISSAN    1.4  2.2  1.0  1.5  0.9  1.2  0.5  1.6  2.0  1.7  1.9  1.0
     RENAULT  -2.5 -1.8 -1.5 -1.6 -1.5 -1.8  0.0 -0.2  2.1 -2.5 -1.7 -1.5
     PORSCHE   1.6  2.5  0.5  2.5  2.5  0.6  2.0  1.8  1.8  1.3 -0.8  0.9
     JAGUAR    3.0  1.4  1.4  1.8  1.2  1.5 -2.4  2.0  0.2  2.3 -1.0 -2.0
     MERCEDES  2.8  1.1  1.2  2.2  1.8  1.9 -2.9  1.3  1.1  2.7  2.1  2.2

Similarities of Automobiles 4 Multiple & Canonical Vector Regression

           Univariate Multiple Regression Statistics for Predicting
                      the Properties from the Dimensions

                  Squared Multiple Correlations and F Tests
                      2 numerator df    8 denominator df

                        Adjusted   Approx 95% CI for RSQ         F
           R-Squared   R-Squared   Lower CL      Upper CL   Statistic   Pr > F
SOPHIST     0.888560    0.860701      0.509         0.962     31.8939   0.0002
EXCITING    0.946784    0.933480      0.735         0.982     71.1650   0.0001
STRONG      0.679001    0.598752      0.077         0.882      8.4611   0.0106
UNCONVEN    0.950630    0.938287      0.752         0.984     77.0204   0.0001
BOLD        0.892560    0.865701      0.523         0.964     33.2302   0.0001
SPORTY      0.817292    0.771614       .            0.937     17.8928   0.0011
FRESH       0.430208    0.287761      0.000         0.759      3.0201   0.1054
ELEGANT     0.810381    0.762976      0.300         0.934     17.0949   0.0013

             Average R-Squared:  Unweighted           = 0.801927
                                 Weighted by Variance = 0.829086

                         Raw Regression Coefficients
                SOPHIST          EXCITING            STRONG          UNCONVEN

 DIM1      -1.617922422      -1.179179565      -0.942544973      -1.294782381
 DIM2      -0.230397693       1.147472790       0.225151404       0.391200134


                   BOLD            SPORTY             FRESH           ELEGANT

 DIM1      -1.211945183       0.393179482      -0.480343595      -1.521561939
 DIM2       0.558754499       1.599190211       0.715110380       0.080032909

                  Prob > |T| for the Regression Coefficients
                    SOPHIST      EXCITING        STRONG      UNCONVEN

         DIM1        0.0001        0.0001        0.0037        0.0001
         DIM2        0.3920        0.0001        0.4619        0.0193


                       BOLD        SPORTY         FRESH       ELEGANT

         DIM1        0.0001        0.1172        0.1529        0.0004
         DIM2        0.0224        0.0004        0.0965        0.8119

Similarities of Automobiles 5 Multiple & Canonical Vector Regression

                        Canonical Correlation Analysis

                                Adjusted       Approx       Squared
                 Canonical      Canonical     Standard     Canonical
                Correlation    Correlation     Error      Correlation
           1      0.997072       0.994933     0.001849      0.994152
           2      0.977344       0.966439     0.014167      0.955200
                                Eigenvalues of INV(E)*H
                                  = CanRsq/(1-CanRsq)

                 Eigenvalue    Difference    Proportion    Cumulative
            1     169.9961      148.6745       0.8886        0.8886
            2      21.3217         .           0.1114        1.0000
                     Test of H0: The canonical correlations in the
                       current row and all that follow are zero

               Likelihood
                  Ratio      Approx F      Num DF      Den DF    Pr > F
          1    0.00026199      7.5977          16           2    0.1224
          2    0.04479951      6.0919           7           2    0.1482

                 Multivariate Statistics and F Approximations

                            S=2    M=2.5    N=-0.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.00026199     7.5977        16         2  0.1224
 Pillai's Trace             1.94935241     9.6221        16         4  0.0205
 Hotelling-Lawley Trace   191.31780781      .            16         0   .
 Roy's Greatest Root      169.99613474    42.4990         8         2  0.0232

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

Similarities of Automobiles 6 Multiple & Canonical Vector Regression

                        Canonical Correlation Analysis

                Raw Canonical Coefficients for the Dimensions
                            CAN1              CAN2

          DIM1      -0.708983668      0.4932275028      Dimension 1
          DIM2      0.6160298908      0.8855044156      Dimension 2

                Raw Canonical Coefficients for the Properties
                                         W1                W2

                 SOPHIST       0.3760465083      -0.230215282
                 EXCITING      0.4657408539      -0.280206977
                 STRONG         -0.16504248      0.7894447402
                 UNCONVEN      -0.280757671      1.4317654465
                 BOLD          -0.002414609       -1.29082502
                 SPORTY         0.236366897      0.4698556865
                 FRESH         -0.104157126      0.5783531863
                 ELEGANT        0.140824761      -0.595270702

            Standardized Canonical Coefficients for the Dimensions
                            CAN1          CAN2

              DIM1       -0.8209        0.5711      Dimension 1
              DIM2        0.5711        0.8209      Dimension 2

            Standardized Canonical Coefficients for the Properties
                                         W1            W2

                     SOPHIST         0.7522       -0.4605
                     EXCITING        0.8284       -0.4984
                     STRONG         -0.2225        1.0645
                     UNCONVEN       -0.4441        2.2650
                     BOLD           -0.0038       -2.0437
                     SPORTY          0.4055        0.8060
                     FRESH          -0.1374        0.7630
                     ELEGANT         0.2758       -1.1660

Similarities of Automobiles 7 Multiple & Canonical Vector Regression

                             Canonical Structure

      Correlations Between the Dimensions and Their Canonical Variables
                            CAN1          CAN2

              DIM1       -0.8209        0.5711      Dimension 1
              DIM2        0.5711        0.8209      Dimension 2

      Correlations Between the Properties and Their Canonical Variables
                                         W1            W2

                     SOPHIST         0.7099       -0.6369
                     EXCITING        0.9745        0.0538
                     STRONG          0.7550       -0.3429
                     UNCONVEN        0.9115       -0.3612
                     BOLD            0.9171       -0.2431
                     SPORTY          0.2765        0.8809
                     FRESH           0.6349        0.1757
                     ELEGANT         0.7622       -0.4937

                 Correlations Between the Dimensions and the
                    Canonical Variables of the Properties
                              W1            W2

              DIM1       -0.8185        0.5581      Dimension 1
              DIM2        0.5694        0.8023      Dimension 2

                 Correlations Between the Properties and the
                    Canonical Variables of the Dimensions
                                       CAN1          CAN2

                     SOPHIST         0.7078       -0.6225
                     EXCITING        0.9716        0.0526
                     STRONG          0.7528       -0.3351
                     UNCONVEN        0.9089       -0.3530
                     BOLD            0.9144       -0.2376
                     SPORTY          0.2757        0.8610
                     FRESH           0.6330        0.1717
                     ELEGANT         0.7600       -0.4825

Similarities of Automobiles 8 Multiple & Canonical Vector Regression

             OBS    _TYPE_        DIM1        DIM2      NAME
               1    B           -1.61792    -0.23040    SOPHIST
               2    B           -1.17918     1.14747    EXCITING
               3    B           -0.94254     0.22515    STRONG
               4    B           -1.29478     0.39120    UNCONVEN
               5    B           -1.21195     0.55875    BOLD
               6    B            0.39318     1.59919    SPORTY
               7    B           -0.48034     0.71511    FRESH
               8    B           -1.52156     0.08003    ELEGANT
               9    RAWSCORE    -0.70898     0.61603    CAN1
              10    RAWSCORE     0.49323     0.88550    CAN2

Similarities of Automobiles 9 Multiple & Canonical Vector Regression

              Plot of DIM2*DIM1$NAME.  Symbol points to label.

           |
       2.0 +
           |
           |
           |
           |
           |                                    > SPORTY
       1.5 +
           |
           |
           |
           |    EXCITING <      > 300ZX        > Mustang
           |                          > Corvette
       1.0 +
           |               > Porsche             > CAN2
           |
    D      |                       > FRESH
    i      |
    m      |             > BOLD > CAN1
    e  0.5 +
    n      |            > UNCONVEN
    s      |
    i      |                 > STRONG
    o      |                                               Escort <
    n      |         > ELEGANT
       0.0 +                              > +
    2      |
           |
           |       > SOPHIST
           |                                               Chevette <
           |
      -0.5 +
           |             > Jaguar                            > Renault
           |
           |
           |
           |                   Cadillac
      -1.0 +   Mercedes <         ^    > Lincoln
           |
           |
           |
           |
           |
      -1.5 +
           ---+-------------+-------------+-------------+-------------+--
             -2            -1             0             1             2

                                     Dimension 1


Similarities of Automobiles 10 Multiple & Canonical Ideal Point Regression

           Univariate Multiple Regression Statistics for Predicting
                      the Properties from the Dimensions

                  Squared Multiple Correlations and F Tests
                      3 numerator df    7 denominator df

                        Adjusted   Approx 95% CI for RSQ         F
           R-Squared   R-Squared   Lower CL      Upper CL   Statistic   Pr > F
SOPHIST     0.904297    0.863281      0.468         0.964     22.0476   0.0006
EXCITING    0.954807    0.935439      0.713         0.983     49.2974   0.0001
STRONG      0.836185    0.765979      0.238         0.937     11.9104   0.0039
UNCONVEN    0.950788    0.929696      0.691         0.982     45.0801   0.0001
SPORTY      0.818378    0.740539      0.192         0.929     10.5138   0.0055
FRESH       0.717142    0.595917       .            0.885      5.9158   0.0247
ELEGANT     0.813190    0.733129      0.179         0.927     10.1571   0.0061

             Average R-Squared:  Unweighted           = 0.856398
                                 Weighted by Variance = 0.865516

                         Raw Regression Coefficients
                SOPHIST          EXCITING            STRONG          UNCONVEN

 SSQ       -0.378478609       0.240337542      -0.806385962      -0.029978405
 DIM1      -1.483756582      -1.264376159      -0.656691433      -1.284155418
 DIM2      -0.266527852       1.170415786       0.148172543       0.388338349


                          SPORTY             FRESH           ELEGANT

          SSQ        0.085274696       1.065991938       0.156609309
          DIM1       0.362950690      -0.858224143      -1.577077942
          DIM2       1.607330666       0.816871633       0.094983078

                  Prob > |T| for the Regression Coefficients
         SOPHIST  EXCITING    STRONG  UNCONVEN    SPORTY     FRESH   ELEGANT

  SSQ     0.3189    0.3017    0.0359    0.8851    0.8437    0.0322    0.7551
  DIM1    0.0004    0.0001    0.0165    0.0001    0.2370    0.0154    0.0019
  DIM2    0.3298    0.0001    0.5303    0.0311    0.0011    0.0254    0.7929

Similarities of Automobiles 11 Multiple & Canonical Ideal Point Regression

                        Canonical Correlation Analysis

                                Adjusted       Approx       Squared
                 Canonical      Canonical     Standard     Canonical
                Correlation    Correlation     Error      Correlation
           1      0.998644       0.997678     0.000857      0.997289
           2      0.972305       0.957385     0.017273      0.945377
           3      0.850433       0.803134     0.087520      0.723237
                                Eigenvalues of INV(E)*H
                                  = CanRsq/(1-CanRsq)

                 Eigenvalue    Difference    Proportion    Cumulative
            1     367.9304      350.6230       0.9486        0.9486
            2      17.3074       14.6942       0.0446        0.9933
            3       2.6132         .           0.0067        1.0000
                     Test of H0: The canonical correlations in the
                       current row and all that follow are zero

               Likelihood
                  Ratio      Approx F      Num DF      Den DF    Pr > F
          1    0.00004098      5.3320          21     3.42157    0.0772
          2    0.01511758      2.3777          12           4    0.2091
          3    0.27676335      1.5679           5           3    0.3775

                 Multivariate Statistics and F Approximations

                            S=3    M=1.5    N=-0.5

 Statistic                     Value          F      Num DF    Den DF  Pr > F
 Wilks' Lambda              0.00004098     5.3320        21   3.42157  0.0772
 Pillai's Trace             2.66590334     3.4198        21         9  0.0309
 Hotelling-Lawley Trace   387.85092700      .            21        -1   .
 Roy's Greatest Root      367.93035062   157.6844         7         3  0.0008

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

Similarities of Automobiles 12 Multiple & Canonical Ideal Point Regression

                        Canonical Correlation Analysis

                Raw Canonical Coefficients for the Dimensions
                            CAN1              CAN2

          SSQ       0.1674760967      -0.195777824
          DIM1      -0.792620122      0.5038140773      Dimension 1
          DIM2      0.5732574272      0.9030614977      Dimension 2

                Raw Canonical Coefficients for the Properties
                                         W1                W2

                 SOPHIST       0.3828503463      0.0092373151
                 EXCITING       0.630070165      -0.838185345
                 STRONG        -0.371723547      0.9706016822
                 UNCONVEN      -0.275007191      -0.617744465
                 SPORTY        0.1396318104      0.8877762372
                 FRESH         -0.169877162      0.6121035522
                 ELEGANT       0.1398628081      0.3227523944

            Standardized Canonical Coefficients for the Dimensions
                            CAN1          CAN2

              SSQ         0.1314       -0.1536
              DIM1       -0.9177        0.5833      Dimension 1
              DIM2        0.5314        0.8372      Dimension 2

            Standardized Canonical Coefficients for the Properties
                                         W1            W2

                     SOPHIST         0.7658        0.0185
                     EXCITING        1.1207       -1.4909
                     STRONG         -0.5012        1.3088
                     UNCONVEN       -0.4350       -0.9772
                     SPORTY          0.2395        1.5229
                     FRESH          -0.2241        0.8075
                     ELEGANT         0.2740        0.6322

Similarities of Automobiles 13 Multiple & Canonical Ideal Point Regression

                             Canonical Structure

      Correlations Between the Dimensions and Their Canonical Variables
                            CAN1          CAN2

              SSQ        -0.4085        0.0571
              DIM1       -0.8490        0.5030      Dimension 1
              DIM2        0.5166        0.8545      Dimension 2

      Correlations Between the Properties and Their Canonical Variables
                                         W1            W2

                     SOPHIST         0.7270       -0.5616
                     EXCITING        0.9719        0.1165
                     STRONG          0.7241       -0.2297
                     UNCONVEN        0.9229       -0.2871
                     SPORTY          0.2251        0.8924
                     FRESH           0.6779        0.1520
                     ELEGANT         0.7901       -0.4391

                 Correlations Between the Dimensions and the
                    Canonical Variables of the Properties
                              W1            W2

              SSQ        -0.4080        0.0555
              DIM1       -0.8478        0.4891      Dimension 1
              DIM2        0.5159        0.8308      Dimension 2

                 Correlations Between the Properties and the
                    Canonical Variables of the Dimensions
                                       CAN1          CAN2

                     SOPHIST         0.7260       -0.5460
                     EXCITING        0.9705        0.1133
                     STRONG          0.7231       -0.2234
                     UNCONVEN        0.9216       -0.2791
                     SPORTY          0.2248        0.8677
                     FRESH           0.6770        0.1478
                     ELEGANT         0.7890       -0.4269

Similarities of Automobiles 14 Multiple & Canonical Ideal Point Regression

             OBS    _TYPE_          DIM1      DIM2      NAME
              1     B            -1.9602    -0.35210    SOPHIST
              2     B             2.6304    -2.43494    EXCITING
              3     B            -0.4072     0.09187    STRONG
              4     B           -21.4180     6.47697    UNCONVEN
              5     B            -2.1281    -9.42443    SPORTY
              6     B             0.4025    -0.38315    FRESH
              7     B             5.0351    -0.30325    ELEGANT
              8     RAWSCORE      2.3664    -1.71146    CAN1
              9     RAWSCORE      1.2867     2.30634    CAN2

Similarities of Automobiles 15 Multiple & Canonical Ideal Point Regression

              Plot of DIM2*DIM1$NAME.  Symbol points to label.

      |
      |
    8 +
      |
      |
      |          > UNCONVEN
    6 +
      |
      |
      |
    4 +
      |
      |
      |                                                            > CAN2
D   2 +
i     |                                                     300ZX
m     |                                               Corvette2< > Mustang
e     |                                                  PorschSTRONG
n   0 +                                                SOPHIST ^Rena>lEscort
s     |                                               Jaguar^<   v ^> Chevevte
i     |                                             Mercedes <Fv>SLincolnELEGA
o     |                                                     Cadillac > CAN1
n  -2 +
      |                                                      EXCITING <
2     |
      |
   -4 +
      |
      |
      |
   -6 +
      |
      |
      |
   -8 +
      |
      |
      |                                                    > SPORTY
  -10 +
      |
      ---+----------+----------+----------+----------+----------+----------+--
        -25        -20        -15        -10        -5          0          5

                                     Dimension 1

NOTE: 9 label characters hidden.