LEEM: Concentration Summary Statistics to Weighted Concentration &...

In AlexanderEast/LEEMR: Exposure estimates from sparse concentration summary statics

Concentration Summary Statistics to Weighted Concentration & Exposure Estimates

Description

A function which creates synthetic lognormal exposure data from a dataframe of summary statistic concentration data, exposure factors, a selected weighting column, and n number of points.

This model is based on methods described in Egeghy & Lorber (2011) , Lorber & Egeghy (2011) , and East et al. (2021).

The series of estimating equations for geometric mean (GM) and geometric standard deviation (GSD) is primarily adapted from Pleil et al. (2014).

Accepted Units: ng/m3, ug/m3, pg/ml, ng/mL, ug/L, ng/L,pg/g,ug/kg, ng/g

Concentration output: ng/m3, ng/L, or ng/g

Exposure output: ng/day

The function returns a list of dataframes:

Summary: Mean, min, 10th percentile, median, 75th percentile, 95th percentile, and max for concentration and exposure across each media, pathway, and chemical entered.

Used Input: The input data for which GM and GSDs could be estimated.

Excluded Input: The input data for which GM and GSDs could not be estimated.

Used Dataset Counts: Counts of used datasets by chemical and media.

Absorption Fractions: The Absorption fractions used.

Raw: Raw data for plotting.

Metadata: Time and date, chemicals, media, n, and seed used in the run.

Usage

LEEM(data, factors, absorption = NULL, wtcol, n, seed = NULL)

The LEEM is used to generate exposure estimates from sparse summary statistics
for any number of chemicals, individuals, and media entered.
Note: LEEM_Template generates an Excel file template with for data and factors arguments.

Arguments

data = A dataframe containing the following columns: Chemical, Media, Units, Sample Size, Min, Max, Median, Mean, SD, GM, GSD, P10, P25, P75, P90, P95, P99. Column names are not case sensitive.

factors = A dataframe with exposure factors for each media and chemical in the data dataframe. Contains the following columns : Chemical, Path, Media, and Absorption, which are not case sensitive. A dataframe will be auto-generated if not entered.

absorption = A dataframe with absorption fractionsfor each media in the data dataframe. Contains the following columns : Path, Media, Individual, and Factor, which are not case sensitive.

wtcol = A character input for the name of the 'weighting' column (often 'sample size').

n = The number of generated concentration and exposure outcomes generated by rlnorm().

seed = Optional input to ensure or characterize variability between runs. Set by default to 12345.

Examples

# Using the Example dataset 'LEEMR_Example_Data':
# sheets  "Water" and "Water Factors" called respectively
# specifying the column "sample size" to weight datasets
# setting n = 1000
# no argument provided for 'seed', defaulting to "12345"

LEEM(
  LEEMR_Example_Data$Water,
  LEEMR_Example_Data$`Water Factors`,
  LEEMR_Example_Data$`Water Absorption`,
     'sample size',
     1000,
     123)

AlexanderEast/LEEMR documentation built on May 21, 2022, 10:10 a.m.