Another approach, not mentioned yet, is to use ace, in the
acepack package. I have used this in an article (with Andy
Gurmankin) coming out soon in Memory and Cognition, which I could
send by email. It isn't obvious to me that this will (or that it
won't) work with a fractional factorial design;
Hi,
The conf.design package should help you handle the experimental design
side of your problem. Depending on your application, it may be unwise to
assume that main effects will be enough, as interactions can often turn
out to be important (at least in my experience with discrete conjoint).
Hope
Hi there
Maybe you can find useful The isoreg {modreg} package which does Monotone
regression. This is probably what you need to model your data.
Nonetheless be aware you need to code properly your design matrix
(use orthogonal polynomial codes)
The difference in using MNP or ordered probit is t
Hi,
try package "MNP" for a starting point - which could be used for
choice-based-conjoint!
And here a paper which show you that a normal Conjoint design
is nothing others than a regression analysis, which could ready easy used
with little bit programming in R.
www.sawtoothsoftware.com/dow
