SAS中每个独立观察的平均___？
Average ___ per distinct observation in sas?

数据步骤中这样的事情

/*sorting*/
proc sort data=have out=have1;
by name;
run;


  data want(drop=new_grade);
  do until(last.name);
  set have1(rename=(grade=new_grade));
  by name;
 grade = sum(new_grade, grade);
end;
run;

或在proc sql中

proc sql;
create table want as 
select name, sum(grade) as grade
from have
group by name;

我为您展示了标准SAS数据集上的示例。 您必须事先按要分组的变量进行排序

proc sort data=sashelp.class out=class(keep=Sex Name Weight);
by sex;
run;

接下来添加一些保留变量（保留“下一个”行中的值）并使用last或first表达式进行条件处理

data res;
set class;
by sex;
retain sum 0 n 0;
sum+Weight;
n+1;

if last.sex then
    do;
        avg=sum/n;
        n=0;
        Sum=0;
    end;
run;

结果

+---------+-----+--------+-------+---+--------------+
|  Name   | Sex | Weight |  sum  | n |     avg      |
+---------+-----+--------+-------+---+--------------+
| Alice   | F   |     84 |    84 | 1 |            . |
| Barbara | F   |     98 |   182 | 2 |            . |
| Carol   | F   |  102.5 | 284.5 | 3 |            . |
| Jane    | F   |   84.5 |   369 | 4 |            . |
| Janet   | F   |  112.5 | 481.5 | 5 |            . |
| Joyce   | F   |   50.5 |   532 | 6 |            . |
| Judy    | F   |     90 |   622 | 7 |            . |
| Louise  | F   |     77 |   699 | 8 |            . |
| Mary    | F   |    112 |     0 | 0 | 90.111111111 |
| Alfred  | M   |  112.5 | 112.5 | 1 |            . |
| Henry   | M   |  102.5 |   215 | 2 |            . |
| James   | M   |     83 |   298 | 3 |            . |
| Jeffrey | M   |     84 |   382 | 4 |            . |
| John    | M   |   99.5 | 481.5 | 5 |            . |
| Philip  | M   |    150 | 631.5 | 6 |            . |
| Robert  | M   |    128 | 759.5 | 7 |            . |
| Ronald  | M   |    133 | 892.5 | 8 |            . |
| Thomas  | M   |     85 | 977.5 | 9 |            . |
| William | M   |    112 |     0 | 0 |       108.95 |
+---------+-----+--------+-------+---+--------------+

您可以通过SAS中的SQL直接执行此操作。

proc sql noprint;
    create table Avg_Grades as
        select Name, mean(grade) as Avg_Grade
        from have
        group by Name
        ;
quit;

此处最好的方法是使用汇总过程，尤其是如果您需要统计数据而不是均值。 请注意，如果您使用CLASS语句，则不必对数据进行预排序。

1. 在PROC MEANS语句上，有几个选项：
   • NWAY-仅包括最高级别，否则SAS也将添加总体总计
   • N / MEAN / NMISS / MEDIAN-要准备的统计数据
   • STACKODS更改输出格式，这通常是大多数人想要的格式。
2. CLASS->分组变量
3. VAR->要分析的变量

4. ODS OUTPUT->将结果发送到名为WANT的表。

    proc means data=sashelp.class NWAY N MEAN NMISS MEDIAN STACKODS; CLASS sex; *this is your grouping variable(s), ie name; VAR height; *this is the variable(s) you want to summarize; ods output summary = want; run;