proc univariate 输出数据集

是 否 +2 论坛币

k人 参与回答

经管之家送您一份

应届毕业生专属福利!

求职就业群

赵安豆老师微信：zhaoandou666

经管之家联合CDA

送您一个全额奖学金名额~ !

立即领取

感谢您参与论坛问题回答

经管之家送您两个论坛币！

+2 论坛币

请教大侠，在proc univariate过程步中我想用用outtable输出描述性结果及正态性检验及t测验、非参数检验到result集，为什么不能输出统计检验的变量呢，总是提醒以下错误：

WARNING: 从未引用过 DROP、KEEP 或 RENAME 列表中的变量 _PROBN_。
WARNING: 从未引用过 DROP、KEEP 或 RENAME 列表中的变量 _PROBT_。
WARNING: 从未引用过 DROP、KEEP 或 RENAME 列表中的变量 _PROBS_。

具体程序如下：
proc univariate mu0=0  all data=new_ori_data_2 outtable=result(keep= GROUP _VAR_ _SUMWGT_ _NMISS_ _MAX_ _MEAN_ _MEDIAN_ _MIN_ _STD_ _PROBN_ _PROBT_ _PROBS_); /*univariate 输出描述性结果及正态性检验及t测验、非参数检验到result集*/
var new_var new_var1;
by group;
run;

我想同时分析两个变量或更多，所以想用outtable输出数据集，但却无法输出正态性检验结果、t测验及秩和检验结果，只能输出描述性统计结果，请各位赐教，谢谢！

扫码加我 拉你入群

请注明：姓名-公司-职位

以便审核进群资格，未注明则拒绝

分享0 收藏1 回帖

关键词：Univariate Variate ATE IAT NIV result

你可以用ods 将结果输出到数据集中的

怎么写的，请高手详细赐教，用ods怎么输出到一个数据集里呢，我都不知数据集名称

Sometimes I use NORMALTEST/NORMAL to request and display tests for normality. But I am not sure if it is relevant to your question. JingJu

Hi,
A little more research on it:
Unfortunately, unlike SAS version 9.3, a lower version does NOT provide all of the statistics you have requested, such as _PROBS_, etc. ODS may be a work-around but OUTPUT statement may be more suitable in this case. The unpleasant is that you have to list the statistics individually in line with the formula SAS requires. I wrote a small macro to work on it. Take an example as per below.
JingJu

1. %macro WriteOptions(statistics, vnum);
   
2.   %do i =1 %to %sysfunc(countw(&statistics));
   
3.     %scan(&statistics, &i)=%do j =1 %to &vnum;%scan(&statistics, &i)&j%str( )%end;
   
4.     %end;%*named by position of variables in VAR statement;
   
5. %mend WriteOptions;
   
6. 
   
7. proc univariate data =sashelp.class noprint;
   
8.   var age weight;
   
9.   by sex notsorted;
   
10.   output out =result %WriteOptions(VAR SUMWGT NMISS MAX MEAN MEDIAN MIN STD PROBN PROBT PROBS,2);
    
11. run;

复制代码