pool_internal: Internal Pool Methods

In rbmi: Reference Based Multiple Imputation

Internal Pool Methods

Description

Dispatches pool methods based upon results object class. See pool() for details.

Usage

pool_internal(results, conf.level, alternative, type, D, ...)

## S3 method for class 'jackknife'
pool_internal(results, conf.level, alternative, type, D, ...)

## S3 method for class 'bootstrap'
pool_internal(
  results,
  conf.level,
  alternative,
  type = c("percentile", "normal"),
  D,
  ...
)

## S3 method for class 'bmlmi'
pool_internal(results, conf.level, alternative, type, D, ...)

## S3 method for class 'rubin'
pool_internal(results, conf.level, alternative, type, D, rubin_df_method, ...)

Arguments

results	a list of results i.e. the x$results element of an analyse object created by analyse()).
conf.level	confidence level of the returned confidence interval. Must be a single number between 0 and 1. Default is 0.95.
alternative	a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater" or "less".
type	a character string of either "percentile" (default) or "normal". Determines what method should be used to calculate the bootstrap confidence intervals. See details. Only used if method_condmean(type = "bootstrap") was specified in the original call to draws().
D	numeric representing the number of imputations between each bootstrap sample in the BMLMI method.
...	not used.
rubin_df_method	a character string specifying the degrees-of-freedom approximation used when pooling with Rubin's rules. "barnard-rubin" (default) uses the Barnard-Rubin (1999) small-sample adjustment, while "rubin" uses the approximation from Rubin (1987). Ignored for other pooling methods.

Value

A list containing the pooled results for a single parameter with elements est (point estimate), ci (confidence interval), se (standard error) and pvalue (p-value). The specific computation is determined by the class of results via S3 dispatch.

rbmi documentation built on Oct. 15, 2026, 9:06 a.m.