This article describes how to go from a Latent Class Choice Model, Hierarchical Bayes, or Ensemble of Models output created in Displayr:
- A document containing a Latent Class or Hierarchical Bayes Choice Model or Ensemble of Models output created in Displayr.
- Select the Latent Class Choice Model output in your document.
- From the object inspector, click SAVE VARIABLE(S) > RLH (Root Likelihood).
- A new variable will appear in the data set.
If the model output contains holdout questions, variables for in-sample and holdout RLH will be created, otherwise, just one RLH variable will be created.
RLH is not able to be saved to a variable for models with simulated data and models created using a CHO data file where respondent IDs were not specified
The RLH for each respondent is the geometric mean of the likelihoods of their tasks, i.e., the kth root of the product of the likelihoods, where k is the number of tasks for the respondent.