The IML manual says that the data step function PCTL will work in PROC
IML. Great, except I can make it work in a useful way.
If x is a 10x1 vector, then
median=pctl(50,x);
doesn't really work the way I would hope it works. It gives me a 10x1
vector in return, the median of each value in the original matrix x.
How non-useful!
It does work properly if I write
median=pctl(50,x[1],x[2],x[3],x[4],x[5],x[6],x[7],x[8],x[9],x[10]);
but this isn't a useful way to do things for general problems where x
can be any size.
So, is there a way to have the PCTL function work in a useful way in
PROC IML?