Methods for estimating uncertainty in factor analytic solutions

By P. Paatero, S. Eberly, S. Brown, G. Norris. 2013

524 citations recorded by the source corpus.

Abstract. The EPA PMF (Environmental Protection Agency positive matrix factorization) version 5.0 and the underlying multilinear engine-executable ME-2 contain three methods for estimating uncertainty in factor analytic models: classical bootstrap (BS), displacement of factor elements (DISP), and bootstrap enhanced by displacement of factor elements (BS-DISP). The goal of these methods is to capture the uncertainty of PMF analyses due to random errors and rotational ambiguity. It is shown that the three methods complement each other: depending on characteristics of the data set, one method may provide better results than the other two. Results are presented using synthetic data sets, including interpretation of diagnostics, and recommendations are given for parameters to report when documenting uncertainty estimates from EPA PMF or ME-2 applications.

Subfields: Atmospheric Science, Automotive Engineering, Industrial and Manufacturing Engineering

View Paper: https://www.semanticscholar.org/paper/0a8049eca6053a5e1e1cad7012b9e4e5f9198c05

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Atmospheric Science

Philip K. Hopke

Citations in the corpus

Review of receptor modeling methods for source apportionment (2016, background): “0) incorporates substantially improved error estimation methods (Paatero et al., 2014) that attempt to estimate the errors arising from the inherent uncertainties in the measured data as well as the rotational ambiguity in the solution.”
Methods for estimating uncertainty in PMF solutions: examples with ambient air and water quality data and guidance on reporting PMF results. (2015, methodology): “S/N details are in the EPA PMF Fundamentals and User Guide (Norris et al., 2014), while details of EEmethods are summarized here and presented fully in Paatero et al. (2014).”