choice-metrics.com

Thank you very much Michiel!

The information you shared is very helpful!

Thinking about how the levels specified will impact relative importance: If the actual data used to inform attribute levels are 22.8%. 57.3% and 67.8%, then setting the levels to be [20%, 45%, 70%] is likely leading to a ...

Hi John,

Related to you response to Q1, could you please clarify whether separate coefficients can be estimated for both attributes B and C if adding constraints? My understanding is that we have to collapse them into a single attribute even if we show them as different attributes.

Thanks!



Q1 ...

When doing a Bayesian efficient design, we sometimes find priors from published literature without paying much attention on the levels of that attribute used in the literature. Here are we assuming that the preference weight (marginal impact of attribute on preference) will not change depending on ...

Dear Prof. Bilemer,

I have a couple of questions regarding the coding of an efficient design:

1) Since the calculation of d-error is dependent on the utility function. Does this mean as long as the utility function and priors specified remain the same, the d-errors will be comparable? i.e., even ...