simData: Simulate Data for Meta-Analysis

In jarbes: Just a Rather Bayesian Evidence Synthesis

Simulate Data for Meta-Analysis

Description

This function simulates aggregated data for a meta-analysis, introducing biased studies at different levels.

Usage

simData(
  mu,
  sigma,
  n.total,
  tau,
  N,
  mu.beta.1,
  mu.beta.2,
  mu.beta.3,
  n.B.1,
  n.B.2,
  n.B.3
)

Arguments

mu	Scalar with the true pooled effect value.
sigma	Scalar with the true intra-study standard deviation.
n.total	A vector with the sample sizes of the studies.
tau	Scalar with the between-studies standard deviation.
N	Scalar with the total number of studies in the meta-analysis.
mu.beta.1	Scalar with the mean bias of studies in the mild bias class.
mu.beta.2	Scalar with the mean bias of studies in the large bias class.
mu.beta.3	Scalar with the mean bias of studies in the extreme bias class.
n.B.1	Scalar with the number of studies in the mild bias class.
n.B.2	Scalar with the number of studies in the large bias class.
n.B.3	Scalar with the number of studies in the extreme bias class.

Value

A dataframe with columns:

TE	Observed study's effect.
seTE	Standard error of the study's effect.
theta	True study's effect.
n.total	Sample size of the study.
B.flag	Bias category: "No B", "Mild B", "Large B", "Extreme B".

Examples

set.seed(123)
simData(mu = 0, sigma = 1, n.total = rep(100, 10), tau = 0.5, N = 10,
        mu.beta.1 = 0.2, mu.beta.2 = 0.5, mu.beta.3 = 1,
        n.B.1 = 2, n.B.2 = 2, n.B.3 = 2)

jarbes documentation built on June 28, 2025, 1:07 a.m.