求助， sas如何一对一匹配

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

求助， sas如何一对一匹配
比如我有两组数据， 一组控制组一组处理组， 要求的是对于处理组的每一个样本，选择控制组中年龄相同，体重差异最小的作为匹配对象，这种应该怎么处理啊？
我使用的方式是proc sql， 具体代码如下
PROC SQL;
CREATE table controls_id
as select
one.id as treat_id,
two.id as control_id,
one.age as treat_age,
two.age as control_age,

one.weight as treat_weight,
two.weight as control_weight,
abs(one.weight - two.weight) as diff
from treat one, control two
where ( one.age=two.age )
group by one.id having calculated diff = min(calculated diff);
run;

麻烦各位大神帮我看一下，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0 回帖

关键词：一对一 calculate proc sql Controls control

PROC SQL;
CREATE table controls_id as select
one.id as treat_id,  two.id as control_id,
one.age as treat_age,  two.age as control_age,
one.weight as treat_weight, two.weight as control_weight,
abs(1-two.weight/one.weight) as diff
/*
我建议改成 差异比例  因为很多研究针对 多个变量建构 最后会变成 总和相加最小的样本
所以应该用比例比较好
*/
from treat one, control two
where ( one.age=two.age )
group by one.id having calculated diff = min(calculated diff);
run;