Skip to contents

Pool a scalar estimate across imputations

Source: R/pool_funs.R
pool_rubin_scalar.Rd

Applies Rubin's rules to combine scalar estimates and their within-imputation variance estimates.

Usage

pool_rubin_scalar(q, u)

Arguments

q

(numeric) Vector of scalar estimates, with one value for each imputed dataset.

u

(numeric) Vector of within-imputation variance estimates corresponding to q.

Value

A named list with the pooled estimate (est), standard error (se), variance (var), and number of imputations (m).

Details

For m imputations, the pooled estimate is the mean of q. The pooled variance combines the mean within-imputation variance with the between-imputation variance, using the usual (1 + 1 / m) multiplier. The returned standard error is the square root of the pooled variance.

See also

pool_z_stat()

Examples

pool_rubin_scalar(
  q = c(0.20, 0.25, 0.15),
  u = c(0.01, 0.016, 0.012)
)
#> $est
#> [1] 0.2
#> 
#> $se
#> [1] 0.1264911
#> 
#> $var
#> [1] 0.016
#> 
#> $m
#> [1] 3
#>