estimate_unit_information_sd: Estimate Unit Information Standard Deviation

In RoBMA: Robust Bayesian Meta-Analyses

Estimate Unit Information Standard Deviation

Description

Estimates the unit information standard deviation (UISD) from sample sizes and standard errors. The UISD is used to scale weakly informative prior distributions for meta-analytic parameters.

Usage

estimate_unit_information_sd(sei, ni)

Arguments

sei	a complete numeric vector of strictly positive standard errors for each study.
ni	a complete numeric vector of strictly positive sample sizes for each study, with the same length as sei.

Details

The unit information standard deviation is computed following Equation 6 in \insertCiterover2021weakly;textualRoBMA:

\text{UISD} = \sqrt{\frac{\sum n_i}{\sum \text{se}_i^{-2}}}

where n_i is the sample size and \text{se}_i is the standard error for each study.

This function is useful when you want to compute the UISD once and pass it to multiple analyses via the prior_unit_information_sd argument in brma() or related functions. This ensures consistent prior scaling across analyses performed on different subsets of the same data.

Value

Returns a single numeric value representing the estimated unit information standard deviation.

References

\insertAllCited

Examples

# Example with simulated data
sei <- c(0.2, 0.3, 0.25, 0.15)
ni  <- c(50, 30, 40, 80)
estimate_unit_information_sd(sei = sei, ni = ni)

RoBMA documentation built on May 7, 2026, 5:08 p.m.