Dear all,

I conducted a discrete choice experiment with 5 attributes (hereafter regrouped under $att), price, and an ASC. Each attribute has between 3 to 5 levels.

Following an estimation using the 'mixlogit' command (written by Pr. Arne Risa Hole), I used the 'wtp' command (also developed by Pr. Hole) to compute willingness-to-pay values for each attribute level:

mixlogit chosen, rand(ASC $att price) group(HHID_card) id(HHID) nrep(500)
wtp ASC $att price, krinsky reps(10000)
matrix wtp1=r(wtp)

I am able to plot the distribution of the individual-level coefficients for each attribute level using:

mixlbeta att1, nrep(500) saving(att1_data) replace
use att1_data, clear
kdensity att1a

However, I would also like to plot the distribution of the individual-level WTP for each attribute level (see example on slides 23-24: https://www.sheffield.ac.uk/polopoly.../file/DCMD.pdf).

Does anyone have an idea how I could do this?

Many thanks in advance,

Laura