auto_corr: Autocorrelation of MCMC samples

In JointAI: Joint Analysis and Imputation of Incomplete Data

Autocorrelation of MCMC samples

Description

This function obtains the autocorrelation of the MCMC samples in an JointAI object via coda::autocorr.diag(). autocorr_plot() visualizes the results using ggplot2.

Usage

auto_corr(object, lags = 0:30, by_chain = TRUE, outcome = 1L,
  start = NULL, end = NULL, thin = NULL)

auto_corr_plot(object, lags = 0:30, by_chain = TRUE, outcome = 1L,
  start = NULL, end = NULL, thin = NULL)

Arguments

object	an object of class JointAI
lags	a numeric vector indicating the lags to consider
by_chain	logical; should the autocorrelation be computed for each chain separately?
outcome	integer; index of the outcome model for which the autocorrelation should be plotted
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.

Value

a matrix or a list of matrix objects if by_chain = TRUE, or a ggplot() object for autocorr_plot().

Examples

fit <- lm_imp(y ~ C1 + C2 + B2, data = wideDF, n.iter = 200)
auto_corr(fit)
auto_corr_plot(fit)

JointAI documentation built on Jan. 30, 2026, 5:07 p.m.