diff --git a/R/mcmc_summary.R b/R/mcmc_summary.R index 4c06d588..2c8a5380 100644 --- a/R/mcmc_summary.R +++ b/R/mcmc_summary.R @@ -331,8 +331,9 @@ summarize_slab = function(fit, component = c("pairwise_samples"), param_names = # n_eff = Var(weight) / composite_MCSE^2, the only calibrated ESS for the # mixture-output row (a raw-chain ESS on the effect inflates in # inclusion-dominated cells). The composite MCSE^2 splits into an inclusion part -# (mu_cond^2 * var_p, with var_p the RB J-chain MCSE^2 of the inclusion -# probability) and a slab part (p_hat^2 * var_mu, from the included-only draws). +# (mu_cond^2 * var_p, with var_p the squared MCSE of the inclusion probability +# estimated from the Rao-Blackwellized inclusion draws) and a slab part +# (p_hat^2 * var_mu, from the included-only draws). # Inputs may be vectors (one entry per edge). derived_weight_ess = function(post_var, p_hat, var_p, mu_cond, var_mu) { # A constant / never-updated inclusion chain contributes no inclusion error. @@ -370,9 +371,9 @@ summarize_pair = function(fit, eap = colMeans(pooled_pw) post_var = apply(pooled_pw, 2, stats::var) - # Inclusion probability and its Monte Carlo error from the RB J-chain; fall - # back to the binary indicator's transition-based MCSE when RB draws are - # unavailable (pre-0.2.0.0 fits). + # Inclusion probability and its Monte Carlo error from the Rao-Blackwellized + # inclusion draws; fall back to the binary indicator's transition-based MCSE + # when these draws are unavailable (pre-0.2.0.0 fits). if(!is.null(fit[[1]][["rb_inclusion_samples"]])) { rb = summarize_manual(fit, array3d = combine_chains(fit, "rb_inclusion_samples")) p_hat = rb$mean @@ -496,8 +497,8 @@ summarize_indicator_compare = function(fit, component = "indicator_samples", par # Summarize one effect with spike-and-slab draws. n_eff is the derived composite # ESS for the model-averaged weight (see derived_weight_ess). The inclusion -# Monte Carlo error comes from the RB J-chain (draws_rb) when available, else -# from the binary indicator's transition-based MCSE. +# Monte Carlo error comes from the Rao-Blackwellized inclusion draws (draws_rb) +# when available, otherwise from the binary indicator's transition-based MCSE. summarize_mixture_effect = function(draws_pw, draws_id, name, draws_rb = NULL) { # Handle case where single-chain extraction returns a vector # (dimension gets dropped when extracting [, , idx] from array with nchains=1) @@ -523,7 +524,8 @@ summarize_mixture_effect = function(draws_pw, draws_id, name, draws_rb = NULL) { mcse_slab = NA_real_ } - ## --- inclusion part: RB J-chain when available, else the binary indicator --- + # Inclusion part: use Rao-Blackwellized inclusion draws when available, + # otherwise use the binary indicator draws. if(!is.null(draws_rb) && any(is.finite(draws_rb))) { rb_array = array(draws_rb, dim = c(niter, nchains, 1L)) pooled_rb = as.vector(rb_array)