rlmc_scaling_low {si4bayesmeta}R Documentation

RLMC adjusted scaling of the study-specific within-study standard deviation in the likelihood: reduced impact of observations

Description

Computes RLMC-adjusted scaled study-specific within-study standard deviations of observations in the likelihood according to Roos et al. (2020). The mean of the scaled observation is assumed to be fixed at the mean of the original observation and only its standard deviation is changed.

Usage

rlmc_scaling_low(sigma, rlmc, hh)

Arguments

sigma

a vector of fixed standard deviations provided in the likelihood

rlmc

the original RLMC

hh

perturbation for RLMC

Details

In this function the reduced impact of observations by increasing their standard deviation is implemented, which we call the 'low' case. Larger scaled sigmai values mean that the observations are made less informative.

Value

Scaled sigma (a vector with within-study standard deviations).

References

Roos, M., Hunanyan, S., Bakka, H., Rue, H., (2020). Sensitivity and identification quantification by a relative latent model complexity perturbation in the Bayesian meta-analysis. Manuscript submitted to Research Synthesis Methods.


[Package si4bayesmeta version 0.1-1 Index]