handles: HANdles Detection Limits when Estimating Sources

In kralljr/handles: HANdles Detection Limits when Estimating Sources

Description

handles helps users estimate sources of PM2.5 with censored data below minimum detection limits (MDL) by first adjusting censored PM2.5 constituent concentrations and then performing source apportionment using Absolute Principal Component Analysis (APCA). Methods for adjusting censored data include a constant proportion of the MDL (e.g. 1/2 MDL), removing constituents with many censored observations, and a likelihood-based method.

Details

Source apportionment

To perform source apportionment, users should have their data as a dataframe where the first column is the date and other columns correspond to constituent concentrations apca

Adjusting censored data below MDLs

To adjust censored data, users can select a substitution method, which substitutes censored data with a constant proportion of the MDL, an exclude method, which excludes constituents with many censored concentrations, and a likelihood-based multiple imputation approach adjust

References

George D. Thurston, John D. Spengler (1985). A quantitative assessment of source contributions to inhalable particulate matter pollution in metropolitan Boston, 19(1) 9-25.

kralljr/handles documentation built on Dec. 6, 2019, 3:39 p.m.