UPCOMING WEBINAR: DIY Data Cleaning & Prep: Super Fast, Super Good This function takes a hierarchical Bayes (Stan) output and produces trace plots of the mean and standard deviation parameters for the distribution from which individual coefficients are sampled. For more information on how to interpret this output, see this blog post, which was written for the closely related MaxDiff analysis.
Note that numeric variables are shown as scaled (this is done to improve model sampling). The prefix "Scaled" is added to numeric variable names. To descale the numeric variable coefficients, multiply them by the multipliers that can be obtained by creating a new R output and typing choice.model$numeric.scaling into the R CODE editor (choice.model needs to be replaced with the name of the choice model output) and clicking calculate. The output will contain a list of multipliers to be used for each numeric variable.
Requirements
A document containing a Hierarchical Bayes output from a choice-based model.
Please note these steps require a Displayr license.
Method
1. Select the Hierarchical Bayes output.
2. From the object inspector, click Data > Diagnostics > Posterior Intervals Plot.
Technical details
Whenever hierarchical Bayes analysis is run with multiple classes, an attempt will be made to match class labels between chains (note that it is often not possible to match class labels). If this succeeds, or if only one chain was specified, one set of mean and standard deviation parameters will be shown for each class. If this attempt is unsuccessful, the trace plots will not be able to be displayed.
Acknowledgments
Uses the rstan R package.
References
McLean, M. W. (2018, July 24). How to Use Hierarchical Bayes for Choice Modeling in Displayr [Blog post]. Accessed from https://www.displayr.com/how-to-hierarchical-bayes-choice-model-displayr/.
Yap, J. (2018, January 16). Checking Convergence When Using Hierarchical Bayes for MaxDiff [Blog post]. Accessed from https://www.displayr.com/convergence-hb-maxdiff/.
Next
How to Do the Statistical Analysis of Choice-Based Conjoint Data