Date:         Tue, 30 Nov 2004 15:48:47 -0500
Reply-To:     Ya Huang <ya.huang@AMYLIN.COM>
Sender:       "SAS(r) Discussion" <SAS-L@LISTSERV.UGA.EDU>
From:         Ya Huang <ya.huang@AMYLIN.COM>
Subject:      Re: Table calculations
Comments: To: Kimberly Ann Fernandes <kafernan@UWATERLOO.CA>

A very good candidate for proc sql:

data xx;
input Center x m y;
cards;
1  0   2   0
2  0   4   3
2  1   2   2
3  0   5   3
3  1   3   1
;

proc sql;
select *,
 max(case when x=0 then y else . end)/max(case when x=1 then y else . end)
as d
from xx
group by center
order by center,x
;
-------

  Center         x         m         y         d
------------------------------------------------
       1         0         2         0         .
       2         0         4         3       1.5
       2         1         2         2       1.5
       3         0         5         3         3
       3         1         3         1         3


Kind regards,

Ya Huang

On Tue, 30 Nov 2004 15:36:29 -0500, K Fernandes <kafernan@UWATERLOO.CA>
wrote:

>Hello,
>
>I have a table of count data that is formatted like this:
>
>Center  x       m       y
>1               0       2       0
>2               0       4       3
>2               1       2       2
>3               0       5       3
>3               1       3       1
>
>...etc.
>
>where x is a treatment, y is the outcome, and m is the total number of
>people who received treatment x.
>
>I would like to add another column to this table, d, that is calculated
>based on the values of y.  I'm not sure what the most efficient way of
>doing this is, since there is only one value of d per given Center
>value. For example, for a given center, say d=(y from row where x=0)/(y
>from row where x=1).  Could someone give me some help as to how I might
>code this?
>
>Another thing which makes this more complicated is that because m
>happens to equal 0 for Center=1, x=1, there is no corresponding row for
>x=1 in the table above.
>
>I hope that made sense.
>Thank you for your help,
>Kim