sum of column

[dharmana01]

Hi,

I came across one dataset  which consist of

state region1 region2 region3

abc   20          30         50

def    40         30          30

ghi    25         35          40

xyz    30         45          20

i need to find out the sum of each region and which region is largest among the 3 regions?

Can some one help me ? 

[PeterClemmensen]

Use proc means 🙂

data have;
input state $ region1 region2 region3;
datalines;
abc 20 30 50
def 40 30 30
ghi 25 35 40
xyz 30 45 20
;

proc means data = have sum;
   var region1-region3;
run;

[dharmana01]

Hi dray,

Thanks for your reply, but i am looking for result into new dataset with the variable state and only the region with larger .

will it can be done in data step itself? 

[PeterClemmensen]

So you want your variable state and the region variable which has the highest sum?

In the example posted, both region 2 and region 3 has sum equal to 140, then what? 🙂

[dharmana01]

Hi dray,

Yes, i want the result with the variable state and the region variable which has the highest sum.

Apologies , observations was randomly given by me .And unfortunate region 2 and 3 sum become equal. you can change any one number either in region 2 and region 3.

Is this possible to do in the datastep itself?

[AbhishekPr]

proc means data=DSN n min max sum;                                                                                                     
run;

You could generate summarization table for above dataset,

It would give you the sum of each region as well as largest among them