ngme_batch_ci: Batch-Means Confidence Intervals from an ngme Fit

In ngme2: Linear Latent Non-Gaussian Models with Flexible Distributions

Batch-Means Confidence Intervals from an ngme Fit

Description

Compute Xi-style batch-means confidence intervals directly from trajectories stored in an 'ngme' object.

Usage

ngme_batch_ci(
  ngme,
  name = "all",
  level = 0.95,
  alpha = 0.501,
  M = NULL,
  N = NULL,
  apply_transform = TRUE,
  drop_burnin = TRUE,
  burnin_iter = 0
)

Arguments

ngme	fitted 'ngme' object with 'store_traj = TRUE' during optimization.
name	either '"all"' (default) to use all parameters jointly, a latent model name, or '"general"' for measurement-noise/fixed-effect block.
level	confidence level.
alpha	stepsize decay exponent in (1/2, 1).
M	number of retained batches (excluding burn-in batch 0).
N	decorrelation constant in the batch boundary formula.
apply_transform	logical; if 'TRUE', inference is first computed in the raw optimization space and then mapped to user scale via post-hoc Delta method (Post-delta).
drop_burnin	logical; if 'TRUE', discard batch 0.
burnin_iter	non-negative integer. Explicitly discard the first 'burnin_iter' optimization iterations before Xi-style batching.

Value

Same structure as [batch_means_ci()] with extra metadata fields 'name' and 'apply_transform'.

ngme2 documentation built on May 20, 2026, 9:10 a.m.