11-30-2014 02:15 PM
I'm working on a discrete choice experiment with 3 factors with 2 levels each = 8 designs. I want to fully know the interaction between the factors and subject preference. I think a 28-set design of two alternatives each would be suitable: I want to confront the 8 designs to each other (8 C 2), like a football league of 8 teams everyone versus everyone. I'm new in SAS and I'm not sure how to develop this procedure. I'm following SAS Technical Support Documents--Marketing Research and I tried this code:
%mktruns(2 ** 3)
%mktex(2 ** 3, n=8)
proc print data=design(obs=8); run;
%choiceff(data=design, /* candidate set of alternatives */
model=class(x1-x3 / sta), /* model with stdzd orthogonal coding */
nsets=28, /* number of choice sets */
maxiter=8, /* maximum number of designs to make */
flags=2, /* 2 alternatives, generic candidates */
options=relative, /* display relative D-efficiency */
beta=zero) /* assumed beta vector, Ho: b=0 */
proc print; by set; id set; run;
%mktdups(generic, data=best, nalts=2, factors=x1-x3)
But I get duplicated sets. I want each factor to appear 7 times in the experiment. I would like to learn some suggestions from you.
Thank you very much in advance.