Skip to contents

Convert between posterior CV (pcov) and the log10-scale concentration SD

Source: R/tidy_extractors.R
pcov_se_conversion.Rd

In the curveR ecosystem the back-calculated concentration is reported on the log10 scale (when is_log_independent = TRUE). se_concentration is the delta-method standard deviation of that log10 concentration, and pcov is the percent coefficient of variation derived from it and then capped:

Usage

pcov_from_se(se)

se_from_pcov(pcov)

Arguments

se

Numeric vector of se_concentration values (log10-scale SD).

pcov

Numeric vector of pcov values (percent).

Value

A numeric vector of the same length.

Details

$$\mathrm{pcov} = \mathrm{se\_concentration} \times \ln(10) \times 100, \quad \text{then capped at } cv\_x\_max.$$

se_concentration is therefore the uncapped modelling-scale SD; pcov is a censored percent. Downstream variance/weight models should consume se_concentration, not pcov (the cap destroys the precision gradient).

These helpers are the single canonical implementation of the relationship. Do not reimplement it elsewhere.