proc sql de 子查询

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

原始数据origin=

id	disposition
001	complete
002	ae
002	complete
002	complete
003	complete
004	death
005	complete
006	death
006	death
006	complete
007	complete
008	sae
009	complete
010	cancel
010	complete
010	complete
011	complete
012	complete
013	ae
013	ae
014	complete
015	complete

标红的都是重复的id，我想去重，然后标记id相同但disposition不同的数。
对，就是刚才提问的升级版，
变成need=

id	disposition	n
001	complete	1
002	ae	2
002	complete	2
003	complete	1
004	death	1
005	complete	1
006	death	2
006	complete	2
007	complete	1
008	sae	1
009	complete	1
010	cancel	2
010	complete	2
011	complete	1
012	complete	1
013	ae	1
014	complete	1
015	complete	1

然后我写的程序是
proc sql noprint;
        create table need as select *,count(*) as n from origin where exists(select distinct * from origin) group by id ;
quit;

但是它没有去重，直接标记了！

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0 回帖

关键词：proc sql ROC sql 子查询 Disposition origin

proc sql;
create table wanted as
select *, count(*) as n from (select distinct * from test)
group by id;
quit;

create table need as select *,count(*) as n from origin where exists(select distinct * from origin) group by id ;     exists需要有连接关键词，a.id=b.id

wwang111 发表于 2016-9-28 14:59
proc sql;
create table wanted as
select *, count(*) as n from (select distinct * from test)

大神，膜拜！