存在两个数据集，如何对一个数据集加weighting，来达到和第一个数据集的变量分布一致。具体例子，如下；数据集A：第一列panelist,第二列GROUPS，第三列INCOME。然后一共100条记录

panelist	Groups	income
10001602	ADULT FAMIL	INCOME 3001-5000 RMB
10001603	ADULT FAMIL	INCOME > 5000 RMB
10002003	YOUNG FAMILY	INCOME > 5000 RMB
10002101	ADULT FAMIL	INCOME > 5000 RMB
10002103	ADULT FAMIL	INCOME > 5000 RMB
10003305	ADULT FAMIL	INCOME < 3000 RMB
10004503	OLDER FAMILY	INCOME > 5000 RMB
10004907	YOUNG FAMILY	INCOME > 5000 RMB
10007606	ADULT FAMIL	INCOME > 5000 RMB
10007610	ADULT FAMIL	INCOME 3001-5000 RMB
10007704	YOUNG FAMILY	INCOME > 5000 RMB
10008118	ADULT FAMIL	INCOME > 5000 RMB
10008306	YOUNG FAMILY	INCOME > 5000 RMB
10008404	OLD SINGLE/COUPLE	INCOME 3001-5000 RMB
10008408	OLDER FAMILY	INCOME 3001-5000 RMB
10008412	ADULT FAMIL	INCOME > 5000 RMB

数据集B也是一样的结构，不同的是，panelist不同。一共80条记录。我现在需要计算出数据集B的每个panelist的一个weight.比如

panelist	Groups	income	weight
10008716	ADULT FAMIL	INCOME < 3000 RMB	1.2
10009106	ADULT FAMIL	INCOME > 5000 RMB	1.3
10009111	OLDER FAMILY	INCOME > 5000 RMB	0.8
10009112	YOUNG FAMILY	INCOME 3001-5000 RMB	2.1
10009604	ADULT FAMIL	INCOME > 5000 RMB	1.5
10009808	OLDER FAMILY	INCOME 3001-5000 RMB	0.3
10010709	ADULT FAMIL	INCOME > 5000 RMB	2.3
10011009	YOUNG FAMILY	INCOME > 5000 RMB	0.1
10011204	OLD SINGLE/COUPLE	INCOME < 3000 RMB	1.5
10011206	ADULT FAMIL	INCOME 3001-5000 RMB	1.2

计算weight的目的是让我
proc freq;
tables groups;
tables income;
run;
出来的结果两个数据集接近，类似。问题就是如何计算这个weight呢？？？？？

跪等高手啊~~~~我有个基本思路就是用每个var里面每个值除以他所在值的总和，然后经过几次迭代之类的。最后模拟出一个最后的freq比较接近的。

求高手啊！！！！！！！！！！！！！！！