How to apply group membership to CBC/HB 4.6.4?

Question

Hi all!

I have an issue with a relatively small sample (N=500) for a new tariff plan launch test.
The new product is much more appreciated with the group of operator's subscribers and some of less liked (but still liked well) by competitors' subs.

It would be ok if there was natural distribution on the operators subs in the sample. But it's not - there are 50% our's to 50% competitors.
So HB sometimes returns the patern of our operator's subs to competitor ones.

Some group memberships, or weights would do the job.
In a least liked scenario I'll splt the HB estimation by 2 groups and estimate 2 different models.
Any hints?

Best regards,
Pavel

Answer 1

Version 4 doesn't really have any options for you.  Version 5 supports covariates so you can include a group membership during estimation.  See http://www.sawtoothsoftware.com/support/technical-papers/hierarchical-bayes-estimation/application-of-covariates-within-sawtooth-software-s-cbc-hb-program-theory-and-practical-example-2009

Comments

Brian, thanks for the clarity.
I hope we will soon have an all in one sawtooth products upgrade

---

We actually do do that now.  I think around in 2010 we started bundling both HB and Latent Class with your CBC license.  If you upgrade your CBC license you'll get HB and Latent Class for free.  We also started offering a 1-user Suite license that includes everything at a significant discount over licensing the individual modules