Summarize cross-biomarker covariance from a Model 2a fit — summarize_cross

Reads posterior draws of the loadings (lambda) and within-biomarker precisions (prec.par) from a Model 2a mcmc.list and returns, per kinetic parameter, the posterior median and 95% credible interval of:

c_p = same-parameter cross-biomarker covariance (\(\lambda_{1,p}\lambda_{2,p}\));
rho_p = the corresponding cross-biomarker correlation.

The per-draw algebra is delegated to the small pure helpers cross_cov_from_loadings() and cross_cor_from_draw_2a(), so this function only handles extraction and summarization.

Usage

summarize_cross_2a(
  mcmc,
  antigens = NULL,
  param_names = NULL,
  probs = c(0.025, 0.5, 0.975)
)

Arguments

mcmc: A coda::mcmc.list from run_mod_2a() (must contain monitored lambda and prec.par nodes).
antigens: Optional length-2 character vector of biomarker labels, used only for the printed pair label.
param_names: Optional length-P character vector of kinetic parameter names. Defaults to the log-scale names.
probs: Quantiles for the credible interval. Default c(0.025, 0.5, 0.975).

Value

A data.frame with one row per kinetic parameter and columns param, pair, cov_med, cov_lo, cov_hi, cor_med, cor_lo, cor_hi.