vnyaga / copuladta Goto Github PK

Hi @VNyaga,

I see that you're using the waic function from the loo package and I just wanted to double check that you're aware of the upcoming update to the loo package (I also sent an email about it back in February). Anyway, the structure of the object returned by waic will change slightly and I think will break the code in your vignette. The new structure of the objects will be:

A named list (of class c("waic", "loo")) with components:

estimates:
A matrix with two columns ("Estimate", "SE") and three rows ("elpd_waic", "p_waic", "waic"). This contains point estimates and standard errors of the expected log pointwise predictive density (elpd_waic), the effective number of parameters (p_waic) and the LOO information criterion waic (which is just -2 * elpd_waic, i.e., converted to deviance scale).

pointwise:
A matrix with three columns (and number of rows equal to the number of observations) containing the pointwise contributions of each of the above measures (elpd_waic, p_waic, waic).

So I think you just need to change the places where you pull results out from the objects. To get elpd_waic it's now object$estimates[1, 1] or object$estimates["elpd_waic", "Estimate"], where object is returned by waic.

I'll send out a follow-up email to my previous one once the loo update is up on CRAN (this week I hope). In the meantime it's now available from the master branch at stan-dev/loo on GitHub.