# Hierarchical multinomial logit with R/JAGS

I am working on a small project where I have to do a Choice Based Conjoint (CBC) analysis. In order to get the part-worths for the different consumers I need to estimate a hierarchical multinomial model. Does anyone know how to do this in R/JAGS, and how to save the estimated coefficients for each consumer in a .csv file?

After a bit og Googling I found the following package which does the trick: http://cran.r-project.org/web/packages/ChoiceModelR/index.html

MCMCpack has implemented a multinomial logit which will probably be considerably faster than one you've implemented yourself in JAGS, presuming that it meets your needs.

MCMCmnl(formula, baseline=NULL, data=NULL,
burnin = 1000, mcmc = 10000, thin = 1,
mcmc.method = c("IndMH", "RWM", "slice"), tune = 1, tdf=6,
verbose = 0, seed = NA, beta.start = NA, b0 = 0, B0 = 0, ...)


More documentation: http://mcmcpack.wustl.edu/

• Hi Zach, thanks for the reference, but MCMCpack does not seem to offer any hierarchical version of multinomial logit. Jun 29, 2012 at 6:43
• Ah indeed, I conveniently read over "hierarchical"
– Zach
Jun 29, 2012 at 17:37