|
Activity Number:
|
405
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Wednesday, August 6, 2008 : 10:30 AM to 12:20 PM
|
|
Sponsor:
|
Section on Statistics and the Environment
|
| Abstract - #302194 |
|
Title:
|
Imputation of Nondetects in Environmental Concentration Measurements
|
|
Author(s):
|
John W. Rogers*+
|
|
Companies:
|
Westat, Inc.
|
|
Address:
|
1650 Research Blvd., Rockville, MD, 20850,
|
|
Keywords:
|
Detection limit ; Imputation ; Non-detect ; Environmental
|
|
Abstract:
|
Nondetects are left-censored measurements generally reported by the laboratory as less than a specified detection limit. Analysis of data with nondetects typically involves substituting an assumed value for the nondetects, such as half the detection limit, or using models that explicitly handle censored data such as survival models using an assumed distribution for the observations. Substitution of an assumed value is a form of imputation. The usefulness of the imputation can be improved by using correlated auxiliary data. This paper presents an approache to imputation of nondetected values based on an assumed distribution for the data and on other correlated data. Examples of correlated data include related but uncensored measurements and the preliminary concentration measurements that are used to determine if the reported concentrations are above or below the detection limit.
|
- The address information is for the authors that have a + after their name.
- Authors who are presenting talks have a * after their name.
Back to the full JSM 2008 program |
