rubin_df: Barnard and Rubin degrees of freedom adjustment

In rbmi: Reference Based Multiple Imputation

Barnard and Rubin degrees of freedom adjustment

Description

Compute degrees of freedom according to the Barnard-Rubin formula.

Usage

rubin_df(v_com, var_b, var_t, M)

Arguments

v_com	Positive number representing the degrees of freedom in the complete-data analysis.
var_b	Between-imputation sample variance of the point estimates across multiply imputed datasets.
var_t	Estimate (according to Rubin's rules) of the variance of the point estimates.
M	Number of imputations (integer larger than 1).

Details

The computation takes into account limit cases where there is no missing data (i.e. the between-variance var_b is zero) or where the complete-data degrees of freedom is set to Inf. Moreover, if v_com is given as NA, the function returns Inf.

Value

Degrees of freedom according to Barnard-Rubin formula. See Barnard-Rubin (1999).

References

Barnard, J. and Rubin, D.B. (1999). Small sample degrees of freedom with multiple imputation. Biometrika, 86, 948-955.

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