SAS Macro Programs: multisummary
$Version: 1.0 (15 Dec 2005)
Michael Friendly
York University
Updated 05/23/2010 13:57:08
Calculate Summary statistics for multiple variables
The MULTISUMMARY macro produces an output data set containg any
of the statistics calculated by PROC SUMMARY for any number of
numeric variables.
The MULTISUMMARY macro is defined with keyword parameters.
The arguments may be listed within parentheses in any order, separated
by commas. For example:
%multisummary(var=x1-x3, stats=n mean std);
- DATA=
-
The name of the input data set [Default:
DATA=_LAST_]
- VAR=
-
The name of the variable to be analyzed [Default:
VAR=_NUMERIC_]
- CLASS=
-
Name(s) of 0 or more class variables
- STATS=
-
List of names of the output statistics to calculate, e.g., MEAN,
MEDIAN, Q1, Q3, etc. Any of the statistic keywords accepted by
PROC SUMMARY.
- OPTIONS=
-
Options for PROC SUMMARY, e.g., nway, missing...
- OUT=
-
The name of the output data set. The observations in this
data set correspond to the combinations of the levels of all
class variables with the set of statistics specified in the
STATS= option.
Example
%include macros(multisummary); *-- or include in an autocall library;
%include data(guerry);
%multisummary(data=guerry,
class=region,
var=Suicides Infants Crime_prop Crime_pers,
stats=Q1 median Q3,
options=missing nway,
out=gstats);
proc print data=gstats;
Obs _statistic_ Region _TYPE_ _FREQ_ Suicides Infants prop pers
1 Q1 1 1 37016 24743 4589 2199
2 median 1 1 37016 24743 4589 2199
3 Q3 1 1 37016 24743 4589 2199
4 Q1 C 1 17 15272 14475 6516 17722
5 median C 1 17 29381 17044 8236 21292
6 Q3 C 1 17 77823 20046 12141 26747
7 Q1 E 1 17 16171 12512 5914 13396
8 median E 1 17 21233 15599 7770 18835
9 Q3 E 1 17 34476 20384 9044 26231
10 Q1 N 1 17 8334 10575 4500 15602
11 median N 1 17 13463 14281 5521 23101
12 Q3 N 1 17 25572 16303 7144 28180
13 Q1 S 1 17 19024 15250 6731 8236
14 median S 1 17 48783 20383 8680 13019
15 Q3 S 1 17 66498 23771 9797 13576
16 Q1 W 1 17 25143 17681 6842 18712
17 median W 1 17 33358 21375 7566 22138
18 Q3 W 1 17 38501 31754 8943 24964
See also
dummy Construct dummy variables for regression models
meanplot Plot means for factorial designs
stat2dat Convert summary dataset to raw data equivalent
sort Generalized dataset sorting by format or statistic
table Construct a grouped frequency table, with recoding
Graphical methods
![[download]](icons/download.gif){kind=icon}
get multisummary.sas)