Cross-correlation of MCMC samples — cross

These functions compute the cross-correlations of the MCMC samples in an JointAI object via coda::crosscorr() and plot them using either the corrplot package or coda::crosscorr.plot().

Usage

cross_corr(object, outcome = NULL, start = NULL, end = NULL,
  thin = NULL)

cross_corr_plot(object, outcome = NULL, start = NULL, end = NULL,
  thin = NULL, type = "corrplot")

Arguments

object: an object of class JointAI
outcome: optional; vector identifying the sub-models whose parameters are included, either by their indices (in the order of the list of model formulas) or their names (LHS of the respective model formula as character string). If NULL (default), all monitored parameters are included. With parameters from several sub-models, the coefficients are named "outcome: variable", otherwise by the variable only.
start: the first iteration of interest (see window.mcmc)
end: the last iteration of interest (see window.mcmc)
thin: thinning interval (integer; see window.mcmc). For example, thin = 1 (default) will keep the MCMC samples from all iterations; thin = 5 would only keep every 5th iteration.
type: character; type of plot to be produced. Either "corrplot" (default) or "coda".

Value

a matrix (or a plot)

Examples

fit <- lm_imp(y ~ C1 + C2 + B2, data = wideDF, n.iter = 200)
cross_corr(fit)
#>              (Intercept)           C1           C2        B21      sigma_y
#> (Intercept)  1.000000000 -0.999337698 -0.009232185  0.1058412 -0.004013081
#> C1          -0.999337698  1.000000000  0.006052573 -0.1407318  0.004398230
#> C2          -0.009232185  0.006052573  1.000000000  0.1199271 -0.006365958
#> B21          0.105841151 -0.140731823  0.119927130  1.0000000 -0.010219705
#> sigma_y     -0.004013081  0.004398230 -0.006365958 -0.0102197  1.000000000
cross_corr_plot(fit, type = "coda")