I would like to build a statistically optimal design for a Discrete Choice Experiment, using the idefix package in R, as described in Traets et al., 2020 (Generating Optimal Designs for Discrete Choice Experiments in R: The idefix Package). Does anyone know how the prior coefficients in par.draws are used to build the efficient design through the Modfed command? Specifically, I'm interested in understanding which are the technical steps that lead from the prior coefficients matrix to the final design.
Thank you in advance,
Alice
Take a look at the attached search for your information. Best wishes David Booth
Dear Alice Stiletto
These links seems to be useful, have a look:
1. https://www.jstatsoft.org/article/view/v096i03
2. https://cran.r-project.org/web/packages/idefix/idefix.pdf
3. http://lab.agr.hokudai.ac.jp/nmvr/02-dce.html
Kind Regards
Qamar Ul Islam
Hi Alice Stiletto
I'm also interested in idefix package in R, but I'm still unclear and curious to know it. Did you find your answer yet? Can you share?
Hi! Late to the answer here, but I've built another R package called cbcTools that is a wrapper around idefix. It was designed to make it easier to generate designs as well as simulate choices and assess power (e.g. sample size requirements) for designs: https://github.com/jhelvy/cbctools
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Statistical Software
- R