I am guessing a little, but is the missing ingredient the LOC function?   So you would need something like:   deaths[yr]=sum(outcome[ loc(time[ ,yr]=1) ,])); ... View more

If you use the 'band' method, it only really makes sense to fill the matrix one row at a time as I have shown.   I think Rick's point is mainly about storage space.  As Paige has pointed out, the full-factorial matrix is going to be huge when you are looking at 2^16 or higher, and if you can manage with a character matrix, it will be reduce the memory requirement by a factor of 8.   I have put both methods in to general modules below, so you can experiment.   proc iml; start FullFactorial( n ); r = 2 ## n; x = j(r, n); pot = 2 ## ( (n-1):0 ); do i = 1 to r; x[i, ] = band( i-1, pot ) > 0; end; return( x ); finish; start FFstr( n ); fmt = cats("binary",char(n),"."); return( putn( t( 0:(2##n - 1) ), fmt) ); finish; a = FullFactorial( 4 ); print a; b = FFstr( 4 ); print b; quit;       ... View more

Use the function EXPANDGRID it does exactly what you want, provided you have 15 factors or less.    x = expandgrid(0:1, 0:1, 0:1); print x; ... View more

Thanks Rick,  that works nicely.   Caution to others using this, the file size is returned as a character string. ... View more

This is how an iterative macro do loop normally operates.  The documentation states "the value of macro-variable changes by the value of increment until the value of macro-variable is outside the range of integers included by start and stop".   So with a default increment of 1, the loop macro variable will be one more than than stop after the loop. ... View more

As a general rule, you will almost never need to mix macros within IML, as a pure IML solution will exist.  For example you can use an IML loop instead of the macro loop, making data set names on the fly using string concatenation:   proc iml; do k=1 to 5; j=k+1; s=J(1,2,0); s[1]=k; s[2]=j; create (cats('s',char(k))) from s; append from s; end; quit; ... View more

It's possible to get there without using loops by making use of the EXPANDGRID function, and it can be generalized to any number of lists by constructing the required EXPANDGRID call in a string and then using CALL EXECUTE.   Here is some code illustrating the smaller problem that you posted.   proc iml; start knap; /* make matrix of all the lists with missing value padding */ lists = {0 26 44 . . ., 0 52 65 78 . ., 0 4 12 20 27 46}; target = 116; /* create binary matrix indicating where the list values are */ values = lists>.; longlist = lists[loc(values)]; /* make one long list */ n = values[ , +]; /* work out where each individual list starts and end in longlist */ m = (1 + cusum(n) - n) || cusum(n); s = rowcat( t( cats( char(m[,1]), ':', char(m[,2]), ',' ))); s = cats( 'e = expandgrid(', s, ');' ); print s; /* show the command s that is about to be executed */ call execute(s); /* all possible combinations of values, one from each list */ c = shape( longlist[e], nrow(e), nrow(lists)); /* now use Rick's trick */ idx = loc(c[,+] = target); if ncol(idx)>0 then print 'Target found :', (c[idx,]); else print 'Target not found'; finish; run knap; quit; ... View more

I think the problem is that you are only saving the results for the last iteration in the loop, since running 'create' will destroy any pre-existing data set.   You need to move the create statement before the loop, so something like: create w var {i w}; do i = 1 to 10; w = ranperm(8); append; end; ... View more