In Sims et al. in prep. (S24) we introduce `information triangle plots' (ITPs), which illustrate the marginal KL divergence (estimated using margarine) between the priors and posteriors of the one- and two-dimensional probability densities of the parameters of the model. ITPs are of general applicability and provide a useful visualisation of how the information in the data is distributed wrt the parameters of the model. Having functionality to generate them in margarine could be a nice feature. If there's interest, I'd be happy to help with this once S24 is finished.
In Sims et al. in prep. (S24) we introduce `information triangle plots' (ITPs), which illustrate the marginal KL divergence (estimated using margarine) between the priors and posteriors of the one- and two-dimensional probability densities of the parameters of the model. ITPs are of general applicability and provide a useful visualisation of how the information in the data is distributed wrt the parameters of the model. Having functionality to generate them in margarine could be a nice feature. If there's interest, I'd be happy to help with this once S24 is finished.