Getting Started

The data set Protein contains the character variable Country and the nine numeric variables representing the food groups. The $14. in the INPUT statement specifies that the variable Country has a length of 14.

The following statements create the distance matrix and display part of it.

proc distance data=Protein out=Dist method=Euclid;


var interval(RedMeat--FruitVeg / std=Std);


id Country;


run;


options ls=120;


proc print data=Dist(Obs=10);


title2 'First 10 observations in the output data set from PROC DISTANCE';


run;

An output SAS data set called Dist that contains the distance matrix is created through the OUT= option. METHOD= Euclid requests that Euclidean (which also is the default) distances should be computed.

The VAR statement lists the variables ( RedMeat ” FruitVeg ) along with their measurement level to be used in the analysis. An interval level of measurement is assigned to those variables. Since variables with large variances tend to have more effect on the proximity measure than those with small variances, each variable is standardized by the STD method to have a mean of 0 and a standard deviation of 1. This is done by adding / with the STD= Std option at the end of the variables list specification.

The ID statement specifies that the variable Country should be copied to the OUT= data set and used to generate names for the distance variables. The distance variables in the output data set are named by the values in the ID variable, and the maximum length for the names of these variables is 14.

There are 25 observations in the input data set; therefore, the output data set Dist contains a 25 by 25 lower triangle matrix.

The PROC PRINT statement displays the first 10 observations in the output data set Dist as shown in Figure 26.1.

{% if main.adsdop %}{% include 'adsenceinline.tpl' %}{% endif %}

Prptein Consumption in Europe


First 10 observations in the output data set from PROC DISTANCE


OBS Country        Albania Austria Belgium Bulgaria Czechoslovakia Denmark E_Germany Finland  France Greece Hungary


1 Albania        0.00000 .        .        .           .          .        .        .       .         .      .


2 Austria        6.12388 0.00000  .        .           .          .        .        .       .         .      .


3 Belgium        5.94109 2.44987 0.00000   .           .          .        .        .       .         .      .


4 Bulgaria       2.76446 4.88331 5.22711  0.00000      .          .        .        .       .         .      .


5 Czechoslovakia 5.13959 2.11498 2.21330  3.94761     0.00000     .        .        .       .         .      .


6 Denmark        6.61002 3.01392 2.52541  6.00803     3.34049    0.00000   .        .       .         .      .


7 E Germany      6.39178 2.56341 2.10211  5.40824     1.87962    2.72112  0.00000   .       .         .      .


8 Finland        5.81458 4.04271 3.45779  5.74882     3.91378    2.61570  3.99426  0.00000  .         .      .


9 France         6.29601 3.58891 2.19329  5.54675     3.36011    3.65772  3.78184  4.56796 0.00000    .      .


10 Greece         4.24495 5.16330 4.69515  3.74849     4.86684    5.59084  5.61496  5.47453 4.54456    0      .


OBS Ireland Italy Netherlands Norway Poland Portugal Romania Spain Sweden Switzerland UK USSR W_Germany Yugoslavia


1    .      .        .         .      .       .       .      .      .        .       .   .      .          .


2    .      .        .         .      .       .       .      .      .        .       .   .      .          .


3    .      .        .         .      .       .       .      .      .        .       .   .      .          .


4    .      .        .         .      .       .       .      .      .        .       .   .      .          .


5    .      .        .         .      .       .       .      .      .        .       .   .      .          .


6    .      .        .         .      .       .       .      .      .        .       .   .      .          .


7    .      .        .         .      .       .       .      .      .        .       .   .      .          .


8    .      .        .         .      .       .       .      .      .        .       .   .      .          .


9    .      .        .         .      .       .       .      .      .        .       .   .      .          .


10    .      .        .         .      .       .       .      .      .        .       .   .      .          .

The following statements produce the tree diagram in Figure 26.2:

proc cluster data=Dist method=Ward outtree=Tree noprint;


id Country;


run;


axis1 order=(0 to 1 by 0.1);


proc tree data=Tree haxis=axis1 horizontal;


height _rsq_;


id Country;


run;

Figure 26.2: Tree Diagram of Clusters versus R-Square Values