Cool model.
This is correct. Both models need to use same y (ie no transformations) or you need to take into account the Jacobian of the transformation. You can add that (log) Jacobian term to your log_lik computation in the generated quantities.
avehtari
2
Related topics
| Topic | Replies | Views | Activity | |
|---|---|---|---|---|
| Loo for hierarchical model with trial-by-trial dependencies | 8 | 329 | April 2, 2025 | |
| How to calculate log_lik in generated quantities of a multivariate regression model | 15 | 4172 | March 9, 2022 | |
| Compare with exact LOO for observations with too high Pareto k diagnostics | 10 | 1624 | January 23, 2020 | |
| Problems in model comparsion with loo-package for a self-written stan model with explanatory variables and hierarchy | 6 | 626 | September 2, 2020 | |
| PSIS-LOO in hierarchical model - only mean log-likelihoods per group work | 8 | 1142 | May 9, 2022 |