At the moment I am analysing survey data using Bayesian estimation in R (bayesm package)
For my analysis, I have included a no-choice option that is coded as a series of zeros. However, I also have a price attribute. This price attribute has three levels but I would like to treat it as a linear attribute to calculate the willingness-to-pay.
The issue is that the zero values of the no-choice options act like 'real' values for the linear price leading to bias in my estimates.
Anyone familiar with this issue and/or workarounds?