So my question is, how do I get from payment estimates to the regression estimates used in the description of priors?
My idea is to do the following:
1. Get min-max estimates for WTP based on previous research
2. Make an educated guess on a min-max parameter value for the price parameter, e.g. -0.1 to -0.0006
3. Use the price parameter to transform the WTP estimates to parameter intervals to construct uniform priors for the other attributes.
I am not sure wether this should be done by a) multiplying the mean price parameter prior with the WTP estimate limits, or allow for wider intervals by b) multiplying low price prior limit with low WTP and correspondingly with the high limits.
Q1) What do you think of this approach? Is there a better strategy? And would you go for maximum uncertainty in parameter priors based on price prior*WTP (ie widest uniform limits under step 3.) or multiply with the mean expected price parameter?
Q2) For the attributes where I do not have a prior - can I leave those at zero or should I rather make a wide guess to avoid that the attributes with priors will dominate the utility functions?
I also have two quick questions regarding the syntax for dummy effects:
My code will look something like this:
U(alt1)=b1[(u,-0.06,-0.03)]*A[-1,0] + b2[(u,0.04,-0.09)]*B[0,1] + b3.dummy[(u,0.09,0.18)|(u,0.06,0.12)]*C[10,20,40] +b4[(u,-0.011,-0.0006)]*price[20,50,100]
Q3) Is it equivalent to write two-level dummies as I have done above with b1 and b2 and to write it out as e.g. b2.dummy[]*B[]?
Q4) With attribute C (minutes) I assume linear effect, but may wish to test this. Therefore I have coded it with dummy. However, the uniform intervals reflect equal per minute disutility for the levels 10 and 20 (compared with 40 as base).
Is this an ok way to model that I expect linearity, but may wish to model the stepwise effects - or is this just equivalent to coding it as a linear effect?
Looking forward to your response!
Thanks
