I am running several models using rstanarm::stan_jm and would like to run various diagnostic models examining residuals from the longitudinal and event submodels to check the functional forms of the relations between predictor variables and residuals. Unlike residuals() for virtually every other model, residuals() for a stan_jm model returns:
Error in residuals.stanmvreg(jointmIAA) : Not currently implemented.
Is there a way to extract from the stanmvreg object the residuals from the two submodels, or alternatively to extract the submodels themselves in a way that would make it possible to extract the residuals?
Thanks to the rstanarm crew for developing these rstanarm functions! I know that I am pushing the limits of your models.