Compute relative standard error (RSE%) for parameter estimates

Calculates the RSE% based on the specified transformation and parameter type. For LogNormal Omega/Sigma, uses sqrt(exp(se^2) - 1) * 100. For other transforms, uses se / |estimate| * 100.

Usage

compute_rse(estimate, se, param_type, transform = "identity")

Arguments

estimate: The parameter estimate(s), can be a vector
se: The standard error(s) of the estimate(s), can be a vector
param_type: Parameter type(s), can be a vector: "Theta", "Omega", or "Sigma"
transform: Transformation type(s), can be a vector: "LogNormal", "AddErr", "Proportional", or "Identity". Defaults to "Identity".

Value

RSE as a percentage (vector)

Examples

if (FALSE) { # \dontrun{
compute_rse(1.5, 0.2, "Theta")
df %>% mutate(rse = compute_rse(estimate, stderr, kind))
} # }