The HPDMDB Procedure

Example 5.1 Running PROC HPDMDB on the Client

This example demonstrates how to run the HPDMDB procedure on the following data set on the client:

```data ex;
input x1 x2 x3 x4 y\$ w f y2;
datalines;
1  2 1 1  m .90 1 0
1  2 1 2  m .91 2 1
1  2 1 3  x .89 1 4
1  2 1 4  x .90 2 4
1  3 1 1  m .91 1 1
1  3 1 2  m .89 2 1
2  3 1 3  x .90 1 5
2  3 1 4  x .89 2 5
3  1 2 1  z .90 1 2
3  1 2 2  z .89 2 2
3  1 2 3  y .90 1 7
3  1 2 4  y .89 2 7
3  4 2 1  z .90 1 3
3  4 2 2  z .89 2 3
4  4 2 3  y .90 1 6
4  4 2 4  y .89 2 6
;
run;
```

When the input data set resides on the client and no PERFORMANCE statement is specified, as in the following example, the client performs all computations:

``` proc hpdmdb data=ex print classout=cout varout=vout;
class x1-x3;
weight w;
var x4 y2;
freq f;
run;
```

Output 5.1.1 shows the summaries of the numeric variables in the data set `ex`.

Output 5.1.1: Summaries of Numeric Variables in `ex` Data Set

Obs NAME NMISS N MIN MAX MEAN STD SKEWNESS KURTOSIS SUM USS CSS
1 x4 0 24 1 4 2.66326 1.06938 -0.05848 -1.19599 57.26 178.80 26.302
2 y2 0 24 0 7 3.57721 2.10658 0.12214 -1.04243 76.91 377.19 102.067

Output 5.1.2 shows the summaries of the classification variables in the data set `ex`.

Output 5.1.2: Summaries of Classification Variables in `ex` Data Set

Obs NAME LEVEL CODE FREQUENCY TYPE CRAW NRAW FREQPERCENT NMISSPERCENT
1 x1 1 0 9 N   1 37.5 37.5
2 x1 2 1 3 N   2 12.5 12.5
3 x1 3 2 9 N   3 37.5 37.5
4 x1 4 3 3 N   4 12.5 12.5
5 x2 1 0 6 N   1 25.0 25.0
6 x2 2 1 6 N   2 25.0 25.0
7 x2 3 2 6 N   3 25.0 25.0
8 x2 4 3 6 N   4 25.0 25.0
9 x3 1 0 12 N   1 50.0 50.0
10 x3 2 1 12 N   2 50.0 50.0