Skip to contents

Fill-in Data Below Limit Of Detection (LOD)

fill_in.Rd

This function fills in data below LOD using the fill-in method by Helsel (1990). It is designed to work with data that should have a normal distribution.

Usage

fill_in(var_to_fill, lod, loq = NULL)

Arguments

var_to_fill

A numeric vector of values to fill.

lod

A numeric vector of the limit of detection (LOD), should be of the same length as var_to_fill.

loq

A numeric vector of the limit of quantification (LOQ), should be of the same length as var_to_fill.

Value

A numeric vector with the original values and where those are below LOD replaced by fill-in values below the limit of detection.

Details

The function flags values that are to be imputed and computes distribution parameters using the NADA::cenros. It then computes fill-in values by performing a random sample between 0 and lod for a normal distribution with the previously computed parameters. Finally, the function replaces values below lod by fill-in values.

References

Helsel, D.R. (1990). Less Than Obvious - Statistical Treatment of Data Below the Detection Limit. Environmental Science & Technology, 24(12), 1766-1774.

Examples

# Example dataset
set.seed(113)
values <- c(0.05, 0.1, 0.2, NA, 0.4, 0.5)
lod <- rep(0.1, length(values))

# Apply fill_in function
imputed_values <- fill_in(values, lod)
print(imputed_values)
#> [1] 0.07635231 0.10000000 0.20000000         NA 0.40000000 0.50000000