Have an idea?

Visit Sawtooth Software Feedback to share your ideas on how we can improve our products.

Can anybody refer a conjoint training data set (constructed or real data) including within and out of sample holdouts?

Dear Sawtooth community,
In the course of my first participation in the modelling challenge I came to learn about a couple of my own conjoint-related blind spots. Now I'd like to keep working on adapting new techniques and refining my processes  but to be able to rate my results independently. To that end I'm curious if anyone is aware of a similar practice data set including holdouts and some build-in challenges (e.g. concept similarity, etc.) which ideally is available free of charge.
Can anybody give me a pointer (or access?!) to corresponding resources?
Thanks!
Alex
asked Jul 29, 2016 by alex.wendland Bronze (2,005 points)
retagged Aug 8, 2016 by alex.wendland
As there seem to be no readily available conjoint training data sets known in the forum here a follow-up question:
Can somebody refer me to a good explanation or tutorial on how I can build my own training set?
I vaguely recall briefly hearing at a conference about the relevant steps:
- define resp groups
- define preferences
- add variance and errors etc.
I'm happy for any pointers! Thanks, Alex

Your answer

Please only use this to answer the original question. Otherwise please use comments.
Your name to display (optional):
Privacy: Your email address will only be used for sending these notifications.
Anti-spam verification:

To avoid this verification in future, please log in or register.
...