Activity Number:
|
353
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 14, 2002 : 2:00 PM to 3:50 PM
|
Sponsor:
|
Section on Statistics & the Environment*
|
Abstract - #301784 |
Title:
|
Measurement Error Models in Chemical Mass Balance Analysis of Pollution Data
|
Author(s):
|
William Christensen*+ and Richard Gunst
|
Affiliation(s):
|
Southern Methodist University and Southern Methodist University
|
Address:
|
P.O. Box 750332, Dallas, Texas, 75275-0332, USA
|
Keywords:
|
Receptor model ; Chemical mass balance model ; Effective-variance weighting
|
Abstract:
|
The chemical mass balance (CMB) equations have been used to apportion observed pollutant concentrations to their various pollution sources. Typical analyses incorporate estimated pollution source profiles, estimated source profile error variances, and error variances associated with the ambient measurement process. Often the CMB model is fit to the data using a weighted least squares algorithm to obtain the effective variance solution. We consider the chemical mass balance model within the framework of the traditional measurement error model (e.g., Fuller, 1987) and compare source contribution estimators with the effective variance estimator.
|