i generated some integer data in minitab ... 10 rows 5 columns
then mtb> rmedian c1-c5, c11 ... gets the median of the row puts it in c11
Row C1 C2 C3 C4 C5 C11
1 18 10 14 12 14 14
2 14 10 13 17 19 14
3 12 18 20 19 20 19
4 11 10 17 19 10 11
5 19 10 18 12 19 18
6 15 18 17 17 16 17
7 13 17 13 15 13 13
8 11 14 14 17 13 14
9 13 11 13 20 12 13
10 15 17 18 17 13 17
now, as per requested, i put some weights on the variables
MTB > let c6=c1*.5
MTB > let c7=c2*1.2
MTB > let c8=c3*.7
MTB > let c9=c4*.3
MTB > let c10=c5*.8
then, did mtb> rmedian c6-c10, c12
here are the data
Row C6 C7 C8 C9 C10 C12
1 9.0 12.0 9.8 3.6 11.2 9.8
2 7.0 12.0 9.1 5.1 15.2 9.1
3 6.0 21.6 14.0 5.7 16.0 14.0
4 5.5 12.0 11.9 5.7 8.0 8.0
5 9.5 12.0 12.6 3.6 15.2 12.0
6 7.5 21.6 11.9 5.1 12.8 11.9
7 6.5 20.4 9.1 4.5 10.4 9.1
8 5.5 16.8 9.8 5.1 10.4 9.8
9 6.5 13.2 9.1 6.0 9.6 9.1
10 7.5 20.4 12.6 5.1 10.4 10.4
does this help?
At 12:54 PM 1/26/01 -0500, Magill, Brett wrote:
>Hi,
>
>Just a practical problem that I have run into here. I have a large data set
>where every case (row) represents a person. Each person belongs to a
>metropolitan area. I want to aggregate some of the individual results into
>metro area statistics. Should be easy... but alas...
=================================================================
Instructions for joining and leaving this list and remarks about
the problem of INAPPROPRIATE MESSAGES are available at
http://jse.stat.ncsu.edu/
=================================================================
